Datasets:

staghado

/

ArXiv-tables

like 2

2309.03451v2

\begin{table} \centering \caption{Sound Types and Characteristics of the data with $Principal Component 1 > 40$.} \begin{tabular}{|P{1.8cm}|c|P{3.5cm}|} \hline Sound Type & Number of Data & Occurrence Date \\ \hline Bearded Seal & 1033 & Mar. or May 2018 \\ \hline Walrus & 9 & Sept. or Oct. 2017 \\ \hline Airgun & 275 & Sept. 2017 \\ \hline Sea ice & 1 & Aug. 25 2017 00:00:00 \\ \hline Whales & 12 & Sept. 11 or 19 2017 \\ \hline Mammal & 7 & Sept. 23 2017 17:00:00 \\ \hline \end{tabular} \label{tab:sound_data} \end{table}

1504.03655v4

\begin{table}[h!] \centering \setlength{\tabcolsep}{2pt} \caption{Relation between various subspaces. \label{tb:relation} } \vspace{-2mm} \begin{tabular}{c|c|c|c|c} \hline \hline Subspace & Evaluation & Orth. & Data Mini-batch & RF Mini-batch \\ \hline $V$ & -- & -- & -- & -- \\ $F_t$ & $f_t(x)$ & \cmark & \cmark & \xmark \\ $G_t$ & $g_t(x)$ & \xmark & \cmark & \xmark \\ $\Gtil_t$ & $\gtil_t(x)$ & \xmark & \cmark & \xmark \\ $H_t$ & $h_t(x)$ & \xmark & \cmark & \cmark \\ \hline \hline \end{tabular} \end{table}

1504.03655v4

\begin{table} \vspace{-4pt} \setlength{\tabcolsep}{10pt} \centering \caption{KCCA results on MNIST 8M (top 50 largest correlations)}\label{table:cca_mnist}\vspace{-4pt} \begin{tabular}{c|c|c|c|c} \hline \multirow{2}{*}{ \# of feat} & \multicolumn{2}{c|}{Random features} & \multicolumn{2}{c}{Nystrom features} \\ \cline{2-5} & corrs. & minutes & corrs. & minutes\\ \hline \hline 256 & 25.2 & 3.2 & 30.4 & 3.0 \\\hline 512 & 30.7 & 7.0 & 35.3 & 5.1 \\\hline 1024 & 35.3 & 13.9 & 38.0 & 10.1 \\\hline 2048 & 38.8 & 54.3 & 41.1 & 27.0 \\\hline 4096 & 41.5 & 186.7 & 42.7 & 71.0 \\\hline \end{tabular}\\ \bigskip \begin{tabular}{c|c|c|c} \hline \multicolumn{2}{c|}{DSGD-KCCA} & \multicolumn{2}{c}{linear CCA}\\ \hline corrs. & minutes & corrs. & minutes\\ \hline \hline 43.5 & 183.2 & 27.4 & 1.1\\\hline \end{tabular} \vspace{-4pt} \end{table}

1504.03655v4

\begin{table}[h!] \centering \setlength{\tabcolsep}{2pt} \caption{Relation between various subspaces. \label{app_tb:relation} } \begin{tabular}{c|c|c|c|c} \hline \hline Subspace & Evaluation & Orth. & Data Mini-batch & RF Mini-batch \\ \hline $V$ & -- & -- & -- & -- \\ $F_t$ & $f_t(x)$ & \cmark & \cmark & \xmark \\ $G_t$ & $g_t(x)$ & \xmark & \cmark & \xmark \\ $\Gtil_t$ & $\gtil_t(x)$ & \xmark & \cmark & \xmark \\ $H_t$ & $h_t(x)$ & \xmark & \cmark & \cmark \\ \hline \hline \end{tabular} \end{table}

1810.02442v1

\begin{table}[tbp] \centering \begin{tabular}{c|c|c|c} \hline Dataset $\#$ & \textsc{w/o L1} & \textsc{DGS} & \textsc{AutoLoss} \\ \hline \hline \emph{1} & .1337 & \textbf{.1019} & .1037 \\ \hline \emph{2} & .1294 & .1035 & \textbf{.1016} \\ \hline \emph{3} & .1318 & .1022 & \bf{.0997} \\ \hline \end{tabular} \vspace{-5pt} \caption{Comparing \textsc{AutoLoss} to other methods when transferring a trained AutoLoss controller for MLP classification to different data distributions.} \label{tab:MLP_transfer} \vspace{-10pt} \end{table}

1810.02442v1

\begin{table} \centering \begin{tabular}{c|c} \hline Feature to drop & \texttt{MSE} \\ \hline \hline (2) normalized gradient magnitude & .086 \\ \hline (3) loss values& .101 \\ \hline (4) validation metrics & .085 \\ \hline None & \textbf{.070} \\ \hline \end{tabular} \caption{The MSE performance on the regression task when some features presented in \S\ref{sec:applications} are ablated.} \label{tab:feature_importance} \end{table}

1808.09545v1

\begin{table}[!htbp] \begin{center} \begin{small} \begin{tabular}{|c|c|c|} \hline TID & A & B \\\hline $t_1$ & $a_1$ & $b_1$ \\\hline $t_2$ & $a_1$ & $b_1$ \\\hline $t_3$ & $a_1$ & $b_2$ \\\hline $t_4$ & $a_1$ & $b_3$ \\\hline $t_5$ & $a_2$ & $b_2$ \\\hline \end{tabular} \end{small} \caption{\label{tb:1f1d}An example of data instance $D$ ($FD: A\rightarrow B$)} \end{center} \end{table}

1808.09545v1

\begin{table} \begin{center} \begin{tabular}{|c|c|}\hline Target attribute set & Covered instance vertex \\\hline $\{AB\}$& $v_{1}$, $v_{2}$, $v_{3}$ (3 vertices)\\\hline $\{A\}$& $v_{1}$, $v_{2}$, $v_{3}$, $v_{4}$ (4 vertices)\\\hline $\{B\}$& $v_{1}$, $v_{2}$, $v_{3}$, $v_{5}$ (4 vertices)\\\hline $\{C\}$& $v_{5}$, $v_{6}$ (2 vertices)\\\hline $\{BC\}$& $v_{5}$, $v_{7}$ (2 vertices)\\\hline \end{tabular} \caption{\label{table:target} An example of target vertex sets} \end{center} \vspace{-0.3in} \end{table}

1508.07096v1

\begin{table}[h!] \centering \begin{tabular}[h!]{|c|c|c|c|c|c|}\hline Algorithms & Sequential & Weight Averaging & Majority Vote & Synchronous Update & Asynchronous Update \\ \hline Error Rate (\%) & 1.08 & 0.98 & 1.04 & 0.97 & 1.06 \\ \hline \end{tabular} \caption{Test error rate using different algorithms for 200 epochs of fine tuning. Weight averaging and majority vote algorithms collect final weights from $7$ independent runs of the standard dropout algorithms. Synchronous update and asynchronous update algorithms combine results from two processes after each input instance. Dropout rate is 50\% for all algorithms. } \end{table}

1908.00868v2

\begin{table} \centering \begin{tabular}{| l || r | } \hline {\bf SVM}&{\bf Ecology}\\ \hline Data point &Species\\ KKT Multiplier & Species Abundance\\ Feature Space & Trait Space\\ Kernel & Niche Overlap\\ Support Vectors & Species that survive in ecosystem\\ \hline \end{tabular} \caption{Conceptual mapping between SVMs and ecology \label{ecotable} } \end{table}

1802.00382v1

\begin{table}[h] \begin{center} \begin{tabular}{|p{1.6 cm}|p{1.2 cm}|p{1.4 cm}|l|l|} \hline \bf Source & \bf Labels & \bf Methods & \bf Rec & \bf F1\\ \hline \hline Gehrmann et al., 2017 & 10 own labels & LR 3-gram & 1.6K & 34.6 \\ \hline Gehrmann et al., 2017 & 10 own labels & CNN & 1.6K & 76 \\ \hline This Paper & 17 ICD-9 & CNN & 5K & 76.2 \\ \hline \end{tabular} \end{center} \caption{Classification of MIMIC clinical notes into labels representing high level phenotype categories (20 epochs for both CNN models)} \label{table:CNN_5k} \end{table}

1802.00382v1

\begin{table}[h] \begin{center} \begin{tabular}{|p{1.7 cm}|p{3 cm}|l|l|} \hline \bf Source & \bf Methods & \bf Recs & \bf F1 \\ \hline \hline This Paper & LSTM& 5k &64.6 \\ This Paper & LSTM-Attention & 5k& 67 \\ \hline This Paper & Hierarchical LSTM-Attention & 5k & 67.6 \\ \hline This Paper & CNN& 5k& 69 \\ This Paper & CNN-Attention& 5k & 72.8 \\ \hline \end{tabular} \end{center} \caption{Classification of MIMIC clinical notes into Level 1 ICD-9 Codes. Evaluation with 17 classes, 5k records, 5 epochs} \label{table:attention} \end{table}

1802.00382v1

\begin{table}[h] \begin{center} \begin{tabular}{|p{1.8 cm}|p {2.6 cm}|l|l|} \hline \bf Source & \bf Methods & \bf N. Rec & \bf F1\\ \hline \hline Perotte et al., 2014& Hierarchal SVM (all codes) & 22K & 39.5 \\ \hline Previous Project Reports\footnotemark& LSTM & 32K & 41.6 \\ \hline This paper &Baseline & 46K & 35 \\ \hline This paper &CNN & 46K & 72.4 \\ \hline \end{tabular} \end{center} \caption{\label{font-table} Classification of MIMIC clinical notes into most common Level 5 ICD-9 Codes} \label{table:top_20} \end{table}

1802.00382v1

\begin{table}[h] \begin{center} \begin{tabular}{|p{1.8 cm}|l|l|l|} \hline \bf Source & \bf Methods & \bf Recs &\bf F1 \\ \hline \hline This Paper & Baseline & 52.6K& 53 \\ This Paper &CNN & 52.6K& 79.7 \\ This Paper &CNN w/ Attention & 52.6K & 78.2 \\ \hline \end{tabular} \end{center} \caption{\label{font-table} Classification of MIMIC clinical notes into 17 Level 1 ICD-9 Codes} \label{table:full_data} \end{table}

1907.13308v2

\begin{table}[!ht] \centering \caption{Outcomes of Holm post-hoc test for AGGLO-2} \small { \begin{tabular}{|l|l|c|c|c|} \hline $ i $ & AGGLO-2 vs. & $ z_i $ & $ p_i $ & $ \cfrac{\alpha}{k - i} $\\ \hline 1 & SVM & 2.9764 & 0.0029 & 0.01 \\ 2 & Decision tree & -1.4174 & 0.1564 & 0.0125 \\ 3 & Naive Bayes & -0.6143 & 0.5390 & 0.0167 \\ 4 & KNN & 0.4725 & 0.6366 & 0.025 \\ 5 & Online GFMM & -0.2835 & 0.7768 & 0.05 \\ \hline \end{tabular} } \label{holm-agglo} \end{table}

1907.13308v2

\begin{table}[!ht] \centering \caption{Outcomes of Holm post-hoc test for incremental learning based GFMM} \small { \begin{tabular}{|l|l|c|c|c|} \hline $ i $ & Online GFMM vs. & $ z_i $ & $ p_i $ & $ \cfrac{\alpha}{k - i} $\\ \hline 1 & SVM & 3.2599 & 0.0011 & 0.01 \\ 2 & Decision tree & -1.1339 & 0.2568 & 0.0125 \\ 3 & KNN & 0.7559 & 0.4497 & 0.0167 \\ 4 & Naive Bayes & -0.3308 & 0.7408 & 0.025 \\ 5 & AGGLO-2 & 0.2835 & 0.7768 & 0.05 \\ \hline \end{tabular} } \label{holm-oln} \end{table}

1907.13308v2

\begin{table}[!ht] \caption{Datasets were used for experiments} \label{table1} \centering \begin{tabular}{|l|L{2cm}|C{1.4cm}|C{1.4cm}|C{1.3cm}|} \hline ID & Dataset & No. samples & No. features & No. classes \\ \hline 1 & Circle & 1000 & 3 & 2 \\ \hline 2 & Complex9 & 3031 & 2 & 9 \\ \hline 3 & Diagnostic Breast Cancer & 569 & 30 & 2 \\ \hline 4 & Glass & 214 & 9 & 6 \\ \hline 5 & Ionosphere & 351 & 34 & 2 \\ \hline 6 & Iris & 150 & 4 & 3 \\ \hline 7 & Ringnorm & 7400 & 20 & 2 \\ \hline 8 & Segmentation & 2310 & 19 & 7 \\ \hline 9 & Spherical\_5\_2 & 250 & 2 & 5 \\ \hline 10 & Spiral & 1000 & 2 & 2 \\ \hline 11 & Thyroid & 215 & 5 & 3 \\ \hline 12 & Twonorm & 7400 & 20 & 2 \\ \hline 13 & Waveform & 5000 & 21 & 3 \\ \hline 14 & Wine & 178 & 13 & 3 \\ \hline 15 & Yeast & 1484 & 8 & 10 \\ \hline 16 & Zelnik6 (Toy dataset) & 238 & 2 & 3 \\ \hline \end{tabular} \end{table}

2201.13299v4

\begin{table}[ht] \caption{Performance of different ablated RES models on the ATOM3D dataset.}\label{table:resr_ab} \vspace{0.3em} \centering \resizebox{\linewidth}{!}{ \begin{tabular}{l|cccc} \toprule %\hline Model & No DirectedLinear & No Interaction & No Equivariance & DW-GNN \\ \midrule Acc \% & 47.3 & 47.7 & 33.0 & \textbf{50.2}\\ \bottomrule % \hline \end{tabular} } \end{table}

1902.10027v3

\begin{table}[t] %\vspace*{-0.1in} \centering \caption{Notations used in \FT approach} \vspace*{-0.1in} \label{table:notation} \begin{tabular}{| c | p{6cm} | } \hline $G$ & A grammar used to generate test inputs \\ \hline $\mathbb{I}_{G}$ & All inputs described by a grammar $G$ \\ \hline $\mathbb{T}_{G}$ & The derivation trees of any input $I \in \mathbb{I}_{G}$ \\ \hline $f_1, f_2$ & Classifiers under test. \\ \hline $J$ & A pre-determined Jaccard Threshold \\ \hline $\tau_G$ & A function $\mathbb{I}_{G} \rightarrow \mathbb{T}_{G}$ which outputs the derivation tree of an input $I \in \mathbb{I}_{G}$ \\ \hline $S$ & The initial input to the directed search. $S$ conforms to grammar $G$ \\ \hline % $Perturb$ & The perturbation function \\ \hline \end{tabular} \vspace*{-0.15in} \end{table}

1902.10027v3

\begin{table}[t] %\vspace*{-0.1in} \centering \caption{Notations used in Evaluation} \vspace*{-0.1in} \label{table:eval-notation} \begin{tabular}{| c | p{6cm} | } \hline \#inputs & Total number of \revise{unique} generated test inputs \\ \hline \#err & Total number of \revise{unique} erroneous inputs \\ \hline $err_r$ & $\frac{\#err}{\#inputs}$ \\ \hline Imp\% & Improvement of $err_r$ of \FT with respect to the $err_r$ of random test\\ \hline \end{tabular} \vspace*{-0.15in} \end{table}

1902.10027v3

\begin{table}[h] \caption{\revise{Jaccard Thresholds}} \centering \begin{tabular}{| c | c | c | c | c | c | c | c |} \hline Grammars & \multicolumn{1}{l|}{A} & \multicolumn{1}{l|}{B} & \multicolumn{1}{l|}{C} & \multicolumn{1}{l|}{D} & \multicolumn{1}{l|}{E} & \multicolumn{1}{l|}{F} \\ \hline $\mathbb{R}$-$\mathbb{A}$ & 0.15 & 0.15 & 0.1 & 0.15 & 0.15 & 0.15 \\ \hline $\mathbb{U}$-$\mathbb{A}$ & 0.15 & 0.15 & 0.1 & 0.15 & 0.1 & 0.1 \\ \hline $\mathbb{R}$-$\mathbb{U}$ & 0.3 & 0.15 & 0.15 & 0.3 & 0.15 & 0.15 \\ \hline \end{tabular} \label{table:thresholds} \end{table}

1902.10027v3

\begin{table}[h] \caption{Sensitivity of \FT w.r.t. Grammars} \vspace*{-0.2in} \label{classifiers} \begin{center} \begin{tabular}{| c | c | c | c | c |} \hline & \multicolumn{2}{c|}{Grammar A - F} & \multicolumn{2}{c|}{$G_{bad}$} \\ \hline & \%unique inps & \%error inps & \%unique inps & \%error inps \\ \hline $\mathbb{R}$-$\mathbb{A}$ & 97\% & 88\% & 52\% & 22\% \\ \hline $\mathbb{U}$-$\mathbb{A}$ & 96\% & 84\% & 49\% & 34\% \\ \hline $\mathbb{R}$-$\mathbb{U}$ & 96\% & 66\% & 51\% & 28\% \\ \hline \end{tabular} \end{center} \label{table:badGrammar} \vspace*{-0.1in} \end{table}

2002.10904v3

\begin{table} \centering \caption{Summary statistics for pretest touches by treatment group.\label{tab:pretest}}. \begin{tabular}{ l c c c c c | c c c c c } \hline & \multicolumn{5}{c}{Experiment 1} & \multicolumn{5}{c}{Experiment 2} \\ & CT & HH & HL & LH & LL & CT & HH & HL & LH & LL\\ \hline Mean & 31.3 & 34.3 & 32.3 & 30.9 & 30.2 & 32.4 & 33.7 & 33.1 & 34.4 & 32.6 \\ SD & 13.4 & 14.6 & 13.6 & 14.3 & 12.6 & 13.6 & 13.1 & 13.5 & 14.2 & 13.6 \\ N & 213 & 150 & 219 & 230 & 257 & 363 & 394 & 416 & 423 & 395 \\ %Shapiro-Wilk & 3e-08 & 4e-4 & 3e-4 & 1e-4 & 8e-05 & 1e-8 & 7e-4 & 5e-8 & 2e-7 & 3e-9\\ \hline \end{tabular} \end{table}

2404.13079v1

\begin{table}[htbp] \centering \caption{Digikala dataset Results} \label{tab:table4} \begin{tabular}{|c|ccc|ccc|} \hline \multirow{3}{*}{Model} & \multicolumn{3}{c|}{2 class} & \multicolumn{3}{c|}{3 class} \\ \cline{2-7} & \multicolumn{1}{c|}{balanced} & \multicolumn{2}{c|}{imbalance} & \multicolumn{1}{c|}{balanced} & \multicolumn{2}{c|}{imbalance} \\ \cline{2-7} & \multicolumn{1}{c|}{accuracy} & \multicolumn{1}{c|}{accuracy} & F1-score & \multicolumn{1}{c|}{accuracy} & \multicolumn{1}{c|}{accuracy} & F1-score \\ \hline ParsBERT & 68 & 87 & 72 & 57 & 62 & 55 \\ ParsBERT + GCN & 70 & 91.1 & 74 & 58 & 63.9 & 55 \\ ParsBERT + RGCN & 70.36 & 91.17 & 74.15 & 58.29 & 63.94 & 55.11\\ \hline \end{tabular} \end{table}

2310.01739v1

\begin{table}[!h] \centering \caption{Asymptotic complexities of various randomized pivoting-based skeleton selection algorithms based on \Cref{algo:sketch_pivot_CUR_general}.} \label{tab:complexity_rand_pivot} \begin{tabular}{c|c|c} \hline Algorithm & Row basis approximator construction (Line 1,2) & Pivoting (Line 3) \\ \hline Rand-LUPP & $O(T_s(l,\Ab))$ & $O(n l^2)$ \\ Rand-LUPP-1piter & $O(T_s(l,\Ab) + \nnz(\Ab) l)$ & $O(n l^2)$ \\ \hline Rand-CPQR & $O(T_s(l,\Ab))$ & $O(n l^2)$ \\ Rand-CPQR-1piter & $O(T_s(l,\Ab) + \nnz(\Ab) l)$ & $O(n l^2)$ \\ \hline RSVD-DEIM & $O\rbr{T_s(l,\Ab) + (m+n)l^2 + \nnz(\Ab) l}$ & $O(nl^2)$ \\ \hline \end{tabular} \end{table}

2310.01739v1

\begin{table}[t] % \begin{wraptable}{hr}{0.65\columnwidth} % \vspace{-1em} \centering \caption{DAC helps FixMatch when the unlabeled data is scarce.} \vspace{-0.5em} \label{table:combining_with_SSL} \begin{tabular}{c|ccc} \hline Number of Unlabeled Data & 5000 & 10000 & 20000 \\ \hline FixMatch & 67.74 & 69.23 & 70.76 \\ FixMatch + DAC ($\lambda=1$) & \textbf{71.24} & \textbf{72.7} & \textbf{74.04} \\ \hline \end{tabular} % \end{wraptable} \end{table}

2307.04012v1

\begin{table}[] \begin{tabular}{r|c|} \cline{2-2} & \textbf{\begin{tabular}[c]{@{}c@{}}CCSD(T) Error\\ kcal/mol\end{tabular}} \\ \hline \multicolumn{1}{|r|}{\textbf{AN1-1x dx}} & 6.44 \\ \hline \multicolumn{1}{|r|}{\textbf{ANI-1x tz}} & 6.98 \\ \hline \multicolumn{1}{|r|}{\textbf{GEOM}} & 16.46 \\ \hline \multicolumn{1}{|r|}{\textbf{Transition1x}} & 11.73 \\ \hline \multicolumn{1}{|r|}{\textbf{QM7-x}} & 8.26 \\ \hline \multicolumn{1}{|r|}{\textbf{Qmugs}} & 14.69 \\ \hline \multicolumn{1}{|r|}{\textbf{ANI-1ccx}} & 13.9 \\ \hline \end{tabular} \caption{The error for the CCSD(T) data after first fitting the six datasets. The ANI-1ccx result is when no previous training is carried out. } \label{tab:ind_err}. \end{table}

2307.04012v1

\begin{table}[] \begin{tabular}{r|c|} \cline{2-2} & \textbf{Number of Structures} \\ \hline \multicolumn{1}{|r|}{\textbf{ANI-1x dz}} & 855,028 \\ \hline \multicolumn{1}{|r|}{\textbf{ANI-1x tz}} & 732,154 \\ \hline \multicolumn{1}{|r|}{\textbf{Qmugs}} & 103,374 \\ \hline \multicolumn{1}{|r|}{\textbf{QM7-x}} & 1,085,249 \\ \hline \multicolumn{1}{|r|}{\textbf{GEOM}} & 122,552 \\ \hline \multicolumn{1}{|r|}{\textbf{Transition-1x}} & 251,095 \\ \hline \multicolumn{1}{|r|}{\textbf{Total}} & 3,149,452 \\ \hline \end{tabular} \caption{The number of structures included from each dataset. } \label{tab:AL_DS} \end{table}

2212.03481v1

\begin{table}[h!] \centering \begin{tabular}{|l|l|l|} \hline \textbf{Station} & \textbf{\# of samples} & \textbf{Execution time} \\ \hline 1 & 84 & 13h 46min \\ \hline 2 & 94 & 24h 26min \\ \hline 3 & 38 & 6h 37min \\ \hline Total & 216 & 44h 49min\\ \hline \end{tabular} \label{train1:time} \caption{Station-wise and total number of samples and execution time of the nf-core HLA typing train} \end{table}

2212.03481v1

\begin{table}[h!] \centering \begin{tabular}{|c|c|c|c|c|c|} \hline \textbf{Station} & \textbf{pre\_run} & \textbf{Execution time} & \textbf{post\_run} & \textbf{Acc}& \textbf{Sens}\\ \hline 1 & 0 / 0MB / 10s & 11h 26m 29s & 8 / 625 MB / 1min 2s & 0.737 & 0.737\\ \hline 2 & 8 / 625 MB / 59s & 12h 47m 24s & 13 / 625 MB / 1m 5s & 0.829 & 0.784\\ \hline 3 & 13 / 625 MB / 57s & 12h 55m 22s & 3 / 654.64MB / 1m 1s & 0.681 & 0.69\\ \hline \textbf{Total} & time 2m 6s & 1d 13h 9m 8s & time 3m 25s & 0.68 & 0.69 \\ \hline \textbf{Central} & 0 / 0MB / 8.817s & 3d 23h 46m 25s & 3 / 654.64MB / 1m 1s & 0.684 & 0.625 \\ \hline \end{tabular} \caption{Execution time and performance of ISIC showcase model at different stations. \textit{pre\_run} and \textit{post\_run} protocols are security protocol steps. Number of files / file size / execution time is reported at each station in the protocol columns. Weighted accuracy (Acc) and weighted sensitivity (Sens) is averaged over all classes and reported from the last epoch at each station.} \label{exp2:time} \end{table}

2212.03481v1

\begin{table}[H] \centering \begin{tabular}{|l|l|} \hline \textbf{Term} & \textbf{Description} \\ \hline $A$ & Algorithm files defined by the User \\ \hline AK & Actinic keratosis \\ \hline BCC & Basal cell carcinoma \\ \hline BKL & Benign keratosis (solar lentigo / seborrheic keratosis / lichen planus-like keratosis) \\ \hline $C_I$ & Container created of image $I$ \\ \hline $CR$ & Container Registry \\ \hline $D_{i}$ & Data (A,Q, and Model / Results) of party $i$ as cargo of the train\\ \hline DF & Dermatofibroma \\ \hline DNN & Deep Neural Network \\ \hline $DS_i$ & Digital Signature of party $i$ \\ \hline $\mathcal{E}_{D}$ & Encrypted value of data $D$\\ \hline GB & Gigabytes \\ \hline GPU & Graphical Processing Unit \\ \hline h & hours \\ \hline HLA & Human Leukocyte Antigen \\ \hline $I$ & Base image \\ \hline $ID_i$ & $ID$ of party $i$ \\ \hline $ID_U$ & Identifier of user $U$ \\ \hline $K$ & Random generated number of length $l$ as session key of the analysis \\ \hline KB & Kilobytes \\ \hline $N$ & Random generated number of length $l$ as session $ID$ of the analysis\\ \hline NV & Melanocytic nevus \\ \hline m & minutes \\ \hline MEL& Melanoma \\ \hline MHC& Major Histocompatibility Complex \\ \hline PDR& Private Docker Registry \\ \hline PHT& Personal Health Train \\ \hline $PK{_i}$ & Public key of the party $i$ \\ \hline $Q$ & Query operated on database defined by the User\\ \hline $R$ & Defined Route of the train defined by the User\\ \hline RE & Result extraction process\\ \hline $S$ & Station \\ \hline SCC & Squamous cell carcinoma \\ \hline $SK{_i}$ & Private key of the party $i$ \\ \hline SMPC & Secure Multi-Party Computation \\ \hline TB & Train building process \\ \hline $U$ & User \\ \hline $UI$ & central User Interface do manage trains and submit algorithms\\ \hline $URI$ & Uniform Resource Identifier \\ \hline VASC & Vascular lesion \\ \hline vCPU & virtual Central Processing Unit \\ \hline \end{tabular}\label{supp:notations} \caption{Notations and abbreviations used in this paper.} \end{table}

1912.10204v1

\begin{table} \vspace{2ex} \resizebox{\columnwidth}{!}{ \begin{tabular}{l | c | c | c} \toprule \textbf{Dataset} & \textbf{\# instances} & \textbf{\# authors} & \textbf{\# texts/author}\\ \midrule \textit{development} & 1000 & 50 & 20\\ \textit{cross-validation} & 3500 & 50 & 70\\ \textit{holdout test} & 500 & 50 & 10\\ \bottomrule \end{tabular} } \centering \caption{Distribution of instances in the different datasets} \label{dist} \end{table}

2109.12567v1

\begin{table}[ht] \begin{tabular}{|l|l|l|l|} \texttt{Metric} & \texttt{c = arrayfun(@(idx)} & \texttt{s = "TestResult" +} & \texttt{String advantage} \\ \hline \texttt{Characters of M-code} & \texttt{78 chars} & \texttt{24 chars} & \texttt{3.25x shorter} \\ \texttt{Duration (sec)} & \texttt{0.01640} & \texttt{0.0003634} & \texttt{45x faster} \\ \texttt{Bytes} & \texttt{129,786} & \texttt{70,096} & \texttt{1.85x smaller} \end{tabular} \caption{Performance comparison of string building for cell and string array} \label{table:StringBuilding} \end{table}

2109.12567v1

\begin{table}[ht] \begin{tabular}{|l|l|l|l|l|} \texttt{Metric} & \texttt{sprintf} & \texttt{{[}num2str(1) ` ' a{]}} & \texttt{1 + " " + a} & \texttt{Advantage over sprintf} \\ \hline \texttt{Chars of M-code} & \texttt{22 chars} & \texttt{18 chars} & \texttt{11 chars} & \texttt{2x as compact} \\ \texttt{Duration (sec)} & \texttt{0.00001375} & \texttt{0.00001227} & \texttt{0.000001693} & \texttt{8.1x faster} \end{tabular} \caption{Comparison of MATLAB text concatenation.} \label{table:Concat} \end{table}

2202.06493v1

\begin{table}[htb] \centering \caption{Dataset for experiments} \small \begin{tabular}{l|rrr} \hline \hline \multicolumn{1}{l}{} & \multicolumn{1}{c}{\begin{tabular}[c]{@{}c@{}}Fashion\\ MNIST\end{tabular}} & \multicolumn{1}{c}{CIFAR10} & \multicolumn{1}{c}{\begin{tabular}[c]{@{}c@{}}Caltech\\ Birds\end{tabular}} \\ \hline Domain & Clothing & Vehicle, Animal & Bird \\ Train data & 60,000 & 50,000 & \multirow{2}{*}{17821} \\ Test data & 10,000 & 10,000 & \\ Classes & 10 & 10 & 200 \\ Clients & 3 & 5 & 10 \\ Samples & 6,400 & 6,400 & 3,200 \\ \hline \hline \end{tabular} \label{tab:dataset} \end{table}

2103.13655v1

\begin{table}[h] \centering % \caption{Overview of the results on the test set after optimization of the Neural Network (NN) and the Structured Deep Kernel Network (SDKN): Cross-correlation (left) and Loss (right).} \renewcommand{\arraystretch}{1.4} \tiny \begin{tabular}{|ll|r|r|l|l|} \hline % \textbf{Cross-Correlation} & & GRU1 & GRU2 & GRU3 \\ \hline \hline Projection & ANN & 0.9989 & 0.8163 & 0.9989 \\ & SDKN & 0.9989 & 0.9989 & 0.9988 \\ \hline Top-Hat & ANN & 0.9992 & 0.9992 & 0.9992 \\ & SDKN & 0.9991 & 0.9992 & 0.9992 \\ \hline Fourier \qquad \qquad & ANN & 0.9992 & 0.9993 & 0.9993 \\ & SDKN & 0.9992 & 0.9993 & 0.9993 \\ \hline \end{tabular} \quad \begin{tabular}{|ll|r|r|l|l|} \hline % \textbf{MSE-Loss} & & GRU1 & GRU2 & GRU3 \\ \hline \hline Projection & ANN & 3.235e-01 & 4.996e+01 & 3.233e-01 \\ & SDKN & 3.253e-01 & 3.261e-01 & 3.368e-01 \\ \hline Top-Hat & ANN & 3.155e-02 & 2.917e-02 & 2.888e-02 \\ & SDKN & 3.222e-02 & 2.989e-02 & 2.893e-02 \\ \hline Fourier \qquad \qquad & ANN & 1.179e-02 & 9.737e-03 & 9.452e-03 \\ & SDKN & 1.177e-02 & 1.007e-02 & 9.587e-03 \\ \hline \end{tabular} \label{tab:results_marius_corr} \end{table}

1904.05961v3

\begin{table}[tb] % \vspace{-.0em} %\small \footnotesize \renewcommand{\arraystretch}{1.3} \caption{Parameters of Datasets} \label{tab:dataset} %\vspace{-1em} \centering \begin{tabular}{r|c|c|c} \hline dataset & size ($|P|$) & dimension ($d$) & \#distinct labels ($L$) \\ \hline Fisher's iris & 150 & 5 & 3 \\ \hline Facebook & 500 & 19 & 4 \\ \hline Pendigits & 7494 & 17 & 10 \\ \hline MNIST & 70000 & 401 & 10 \\ \hline HAR & 10299 & 562 & 6 \\ \hline \end{tabular} %\vspace{-.5em} \end{table}

1904.05961v3

\begin{table}[t] { %\small \footnotesize \renewcommand{\arraystretch}{1.3} \caption{Average Running Time (sec) (`FP': farthest point, `NS': nonuniform sampling, `US': uniform sampling, `RS': RCC-kmeans, `RN': RCC-kmedian) } \label{tab:time} %\vspace{-1em} \centerline{ \begin{tabular}{r|c|c|c|c|c} \hline algorithm & Fisher & Facebook & Pendigits & MNIST & HAR \\ \hline FP %farthest point & 1.62 & 3.00 & 2.53 & 21.69 & 25.92 \\ \hline NS %nonuniform samp. & 0.019 & 0.027 & 0.095 & 7.42 & 0.69 \\ \hline US %uniform samp. & 2.10e-04 & 4.60e-04 & 3.80e-04 & 0.01 & 0.0013 \\ \hline RS %RCC-kmeans & 0.0083 & 0.011 & 0.042 & 18.76 & 1.46 \\ \hline RN %RCC-kmedian & 0.028 & 0.30 & 0.40 & 100.64 & 12.39 \\ \hline \end{tabular} } } \vspace{.5em} \end{table}

1904.05961v3

\begin{table}[t] {%\color{blue} %\small \footnotesize \renewcommand{\arraystretch}{1.3} \caption{Average Running Time (sec) } \label{tab:time, distributed} %\vspace{-1em} \centerline{ \begin{tabular}{r|c|c} \hline algorithm & MNIST & HAR \\ \hline CDCC & 13.84 & 1.55 \\ \hline DRCC & 31.66 & 2.42 \\ \hline \end{tabular} } } \vspace{.5em} \end{table}

1904.05961v3

\begin{table}[tb] % \vspace{-.0em} % \footnotesize % \renewcommand{\arraystretch}{1.3} % \caption{Average Running Time (sec) } \label{tab:time} % \vspace{-.5em} % \centering % \begin{tabular}{r|c|c|c|c} % \hline % algorithm & Fisher's iris & Facebook & Pendigits & MNIST \\ % \hline % farthest point & $0.02$ & $0.04$ & $0.06$ & $7.27$\\ % \hline % nonuniform sampling & $0.06$ & $0.03$ & $0.19$ & $4.34$ \\ % \hline % uniform sampling & $0.001$ & $0.001$ & $0.003$ & $0.01$ \\ % \hline % RCC-kmeans & $0.04$ & $0.02$ & $0.13$ & $13.59$ \\ % \hline % RCC-kmedian & $0.05$ & $0.48$ & $1.28$ & $123.82$ \\ % \hline % \end{tabular} % \vspace{-1em} % \end{table}

1904.05961v3

\begin{table}[tb] %\vspace{-.0em} \color{blue} \footnotesize \renewcommand{\arraystretch}{1.3} \caption{Error bound $\epsilon$ for MEB ($z=1$)} \label{tab:epsilon, z=1} %\vspace{-.5em} \centering \begin{tabular}{r|c|c|c} \hline dataset & coreset size & max relative error & $\epsilon$ \\ \hline Fisher's iris & 5 (20) & 0.0041 & 0.4127 \\ \hline Facebook & 20 (80) & 0.0044 & 49.3484 \\ \hline Pendigits & 40 (400) & 0.0027 & 691.345 \\ \hline MNIST & 400 & 8.1756 & 189130 \\ \hline HARS & 400 & & \\ \hline %LR & $\dist(p_l, p_n^T x_{1:d-1} + x_d)^2$ & weighted sum \\ %\hline \end{tabular} %\vspace{-.5em} \end{table}

1904.05961v3

\begin{table}[tb] %\vspace{-.0em} \color{blue} \footnotesize \renewcommand{\arraystretch}{1.3} \caption{Error bound $\epsilon$ for MEB ($z=2$)} \label{tab:epsilon, z=2} %\vspace{-.5em} \centering \begin{tabular}{r|c|c|c} \hline dataset & coreset size & max relative error & $\epsilon$ \\ \hline Fisher's iris & 5 (20) & 0.0097 & 0.6424 \\ \hline Facebook & 20 (80) & 0.0345 & 7.0248 \\ \hline Pendigits & 40 (400) & 0.0023 & 26.2934 \\ \hline MNIST & 400 & 8.1726 & 434.8897 \\ \hline (new data) & & & \\ \hline %LR & $\dist(p_l, p_n^T x_{1:d-1} + x_d)^2$ & weighted sum \\ %\hline \end{tabular} %\vspace{-.5em} \end{table}

1904.05961v3

\begin{table}[t] { %\small \footnotesize %\color{blue} \renewcommand{\arraystretch}{1.3} \caption{Original Machine Learning Performance (cost for MEB, $k$-means, PCA; accuracy for SVM and NN)} \label{tab:original cost} %\vspace{-1em} \centerline{ \begin{tabular}{r|c|c|c|c|c} \hline problem & Fisher's & Facebook & Pendigits & MNIST & HAR\\ \hline MEB & 2.05 & 7.65 & 18.06 & 90.43 & 60.26 \\ \hline $k$-means & 84.54 & 853.67 & 2.00e+05 & 4.98e+07 & 2.86e+06 \\ \hline PCA & 1.94 & 197.19 & 910.79 & 1.73e+04 & 4.46e+05 \\ \hline SVM/NN\footnotemark & 100\% & 89.36\% & 99.33\% & 87.01\% & 78.01\% \\ %\hline %NN & & & & & \\ \hline \end{tabular} } } \vspace{.5em} \end{table}

1904.05961v3

\begin{table}[t] { %\small \footnotesize \renewcommand{\arraystretch}{1.3} \caption{Average Running Time (sec) (`FP': farthest point, `NS': nonuniform sampling, `US': uniform sampling, `RS': RCC-kmeans, `RN': RCC-kmedian) } \label{tab:time} %\vspace{-1em} \centerline{ \begin{tabular}{r|c|c|c|c|c} \hline algorithm & Fisher & Facebook & Pendigits & MNIST & HAR \\ \hline FP %farthest point & 1.62 & 3.00 & 2.53 & 21.69 & 25.92 \\ \hline NS %nonuniform samp. & 0.019 & 0.027 & 0.095 & 7.42 & 0.69 \\ \hline US %uniform samp. & 2.10e-04 & 4.60e-04 & 3.80e-04 & 0.01 & 0.0013 \\ \hline RS %RCC-kmeans & 0.0083 & 0.011 & 0.042 & 18.76 & 1.46 \\ \hline RN %RCC-kmedian & 0.028 & 0.30 & 0.40 & 100.64 & 12.39 \\ \hline \end{tabular} } } \vspace{.5em} \end{table}

1904.05961v3

\begin{table}[tb] % \vspace{-.0em} % \footnotesize % \renewcommand{\arraystretch}{1.3} % \caption{Average Running Time (sec) } \label{tab:time} % \vspace{-.5em} % \centering % \begin{tabular}{r|c|c|c|c} % \hline % algorithm & Fisher's iris & Facebook & Pendigits & MNIST \\ % \hline % farthest point & $0.02$ & $0.04$ & $0.06$ & $7.27$\\ % \hline % nonuniform sampling & $0.06$ & $0.03$ & $0.19$ & $4.34$ \\ % \hline % uniform sampling & $0.001$ & $0.001$ & $0.003$ & $0.01$ \\ % \hline % RCC-kmeans & $0.04$ & $0.02$ & $0.13$ & $13.59$ \\ % \hline % RCC-kmedian & $0.05$ & $0.48$ & $1.28$ & $123.82$ \\ % \hline % \end{tabular} % \vspace{-1em} % \end{table}

2301.01542v1

\begin{table}[t] \caption{Average test accuracy across clients for different datasets in the settings when $N_{\text{hist}} /N = 50\%$.} \label{tab:ratio_estimation_exp} % \vskip 0.15in \begin{center} \begin{small} \begin{sc} \begin{tabular}{ l | c c c c } \toprule \textbf{Dataset} & $D$ & $G$ & $B$ & $d$ \\ \midrule Synthetic & $1.9$ & $0.4$ & $0.7$ & $21$ \\ CIFAR-10 & $1.0$ & $5.5$ & $2.3$ & $3,353,034$ \\ CIFAR-100 & $1.0$ & $4.7$ & $4.6$ & $3,537,444$ \\ FEMNIST & $5.9$ & $12.9$ & $3.5$ & $867,390$ \\ Shakespeare & $2.6$ & $1.4$ & $6.1$ & $226,180$ \\ \bottomrule \end{tabular} \end{sc} \end{small} \end{center} % \vskip -0.1in \end{table}

1908.04909v1

\begin{table}[h] \begin{center} \begin{tabular}{|l|r|r|r|r|r|r|r|}\hline & TP & FP & FN & TN & ACC & MCC & FPR \\\hline \textbf{model A} & 900 & 500 & 100 & 8500 & 94.0\% & 0.73 & 5.6\% \\\hline \textbf{model B} & 350 & 100 & 650 & 8900 & 92.5\% & 0.49 & 1.1\% \\\hline \end{tabular} \end{center} \caption[Two candidate models]% {Performance of models A and B. TP is true positives; FP is false positives; FN is false negatives; TN is true negatives; ACC is accuracy; FPR is false positive rate. Compared to model B, model A has better MCC, but worse FPR.} \label{ex_data_tab} %\vspace{-20pt} \end{table}

1908.04909v1

\begin{table}[h] \begin{center} \begin{tabular}{ | c |c | c | } \hline \textbf{Target} & \textbf{Predicted False} & \textbf{Predicted True} \\ \hline \textbf{False} & 146,956 & 5,562 \\ \hline \textbf{True} & 22,963 & 10,721 \\ \hline \end{tabular} \end{center} \caption{Confusion Matrix - Validation Data - Default Model} \label{donors_tab2} % \vspace{-20pt} \end{table}

1908.04909v1

\begin{table}[h] \begin{center} \begin{tabular}{ | c | c | c | } \hline \textbf{Target} & \textbf{Predicted False} & \textbf{Predicted True} \\ \hline \textbf{False} & 147,417 & 5,101\\ \hline \textbf{True} & 13,650 & 20,034 \\ \hline \end{tabular} \end{center} \caption{Confusion Matrix - Validation Data - "Best" Model} \label{donors_tab3} % \vspace{-20pt} \end{table}

2304.14853v1

\begin{table}[H] \centering \caption{$p$-values for Sleep States using TDA} \begin{tabular}{| c | c | c | c | c |} \hline \textbf{EEG Band} & NREM1 & NREM2 & NREM3 & REM \\ \hline Delta Band & 0.000 & 0.001 & 0.000 & 0.040 \\ \hline Theta Band & 0.000 & 0.000 & 0.000 & 0.000 \\ \hline Alpha Band & 0.000 & 0.000 & 0.000 & 0.000 \\ \hline Beta Band & 0.000 & 0.000 & 0.000 & 0.000 \\ \hline Gamma Band & 0.000 & 0.000 & 0.000 & 0.000 \\ \hline \end{tabular} \label{tab:sleep_state_p_value_pl} \end{table}

2405.17731v1

\begin{table}[htbp] \centering \caption{Data Loading times for all databases across the five datasets (time in seconds)} \label{tab:loadtime} \begin{tabular}{|c|c|c|c|c|c|} \hline Databases & SF1 & SF2 & SF3 & SF4 & SF5 \\ \hline PostgreSQL & 37s & 375s & 857s & 1089s & 1481s\\ \hline MongoDB & 90s & 1250s & 1701s & 2275s & 2810s \\ \hline ArangoDB & 295s & 2249s & 3964s & 12169s & 15162s \\ \hline Redis & 1495s & 3245s & 5023s & 7748s & 10289s\\ \hline Apache Kudu & 42s & 95s & 146s & 192s & 240s\\ \hline \end{tabular} \end{table}

1903.09493v1

\begin{table}[ht] \centering \begin{tabular}{c|c} \hline \textbf{Reference} & \textbf{Paper} \\ & \textbf{number} \\ \cline{1-2} \cite{Dwork:12} & [1] \\ \cite{Kleinberg:16} & [2] \\ \cite{Berk:17} & [3] \\ \cite{Hardt:16} & [4] \\ \cite{Corbett:17} & [5] \\ \cite{Simoiu:17} & [6] \\ \cite{Chouldechova:16} & [7] \\ \cite{Zafar:17} & [8] \\ \cite{Kusner:18} & [9] \\ \cite{Dieterich:16} & [10] \\ \cite{Zafar2:17} & [11] \\ \cite{Binns:18} & [12] \\ \hline \end{tabular} \caption{References} \label{Table:2} \end{table}

2309.04210v1

\begin{table} \renewcommand{\arraystretch}{1.3} \caption{Comparison of all three observers, showing the mean and standard deviation of the rms voltage error $\obserrrms$ (in $mV$) across twenty trials.} \label{table:results_full} \centering \begin{tabular}{|c|c|c|c|c|} \hline & Centralized & Distributed & $\numredundant=3$ & $\numredundant=9$\\ \hline\hline Mean & 1.15 & 0.0788 & 0.0280 & 0.0241 \\ \hline Standard Deviation & 0.14 & 0.017 & 0.0056 & 0.0030 \\ \hline \end{tabular} \end{table}

1604.02608v1

\begin{table}[!ht] \caption{Data intensive users supported by the OSDC} \begin{center} \begin{tabular}{|l | r|} \hline \# core hours during month & \# of users \\ \hline % 10,000 & 435 \\ \hline 20,000 & 120 \\ \hline 50,000 & 34 \\ \hline 100,000 & 23 \\ \hline 200,000 & 5 \\ \hline \end{tabular} \end{center} \label{tab:users} \footnotesize{The estimated cost of 100,000 core hours on a commercial cloud service provider like AWS is \$40,000 per month.} \end{table}

2101.04025v2

\begin{table}[h] \renewcommand{\arraystretch}{1.3} \caption{Serverless Fit Times and Costs with 1024 MB Memory and Per-Sample-Split Scaling (Mean, Min \& Max in 100 Runs).} \label{table_timing_example} \centering \begin{tabular}{l || c | c | c} \hline & \bfseries Mean & \bfseries Min & \bfseries Max \\ \hline\hline \bfseries Fit Time (s) & 19.82 & 19.53 & 21.49 \\ \bfseries Billed Duration (GB-s) & 3515.36 & 3492.01 & 3571.42 \\ \bfseries Avg.\ Duration per Invocation (s) & 17.16 & 17.05 & 17.44 \\ \bfseries Total Response Time (s) & 19.09 & 18.81 & 20.76 \\ \hline \end{tabular} \end{table}

2005.03197v4

\begin{table}[htb!] %\fontsize{7}{7}\selectfont \caption{(MFC Results) Vanilla HAC and FHAC (Ours)}\label{table1} \begin{center} \tabcolsep=0.10cm \resizebox{0.42\textwidth}{!}{ \begin{tabular}{|c|c|c|c|c|c|c|c|}% \hline \textbf{Dataset} & \textbf{Linkage} & \textbf{$\theta_1$} & \textbf{$\theta_2$} & \textbf{MFC (Vanilla)} & \textbf{MFC (FHAC)}\\ \hline \hline \texttt{census} & Average & 0.001 & 0.001 & \textbf{0.0853} & \textbf{ 0.0853}\\ \hline \texttt{census} & Complete & 0.001 & 0.001 & 0.1806 & \textbf{0.0853}\\ \hline \texttt{census} & Single & 0.001 & 0.65 & 1.248 & \textbf{0.752}\\ \hline \texttt{creditcard} & Average & 0.00005 & 0.005 & 1.2439 & \textbf{1.0}\\ \hline \texttt{creditcard} & Complete & 0.00005 & 0.005 & 0.744 & \textbf{0.60}\\ \hline \texttt{creditcard} & Single & 0.0075 & 0.5 & 1.5 & \textbf{ 0.778}\\ \hline \texttt{bank} & Average & 0.0001 & 0.05 & 1.332 & \textbf{0.6679}\\ \hline \texttt{bank} & Complete & 0.5 & 0.05 & 1.332 & \textbf{0.4457}\\ \hline \texttt{bank} & Single & 0.001 & 0.65 & 1.332 & \textbf{0.6653}\\ \hline \end{tabular}} \end{center} \end{table}

2005.03197v4

\begin{table}[htb!] \caption{(Balance Results) Vanilla HAC and FHAC (Ours)}\label{table2} \begin{center} \tabcolsep=0.09cm \resizebox{0.42\textwidth}{!}{ \begin{tabular}{|c|c|c|c|c|c|} \hline \textbf{Dataset} & \textbf{Linkage} & $\theta_1$ & $\theta_2$ & \textbf{Balance(Vanilla)} & \textbf{Balance(FHAC)}\\ \hline \hline \texttt{census} & Average & 0.001 & 0.001 & \textbf{0.8865} & \textbf{0.8865}\\ \hline \texttt{census} & Complete & 0.001 & 0.001 & 0.7599 & \textbf{0.8865}\\ \hline \texttt{census} & Single & 0.0005 & 1.0 & 0.0 & \textbf{0.9385}\\ \hline \texttt{creditcard} & Average & 0.05 & 0.1 & 0.0 & \textbf{0.427}\\ \hline \texttt{creditcard} & Complete & 0.0005 & 1.0 & 0.0 & \textbf{0.7105}\\ \hline \texttt{creditcard} & Single & 0.0005 & 1.0 & 0.0 & \textbf{0.7959}\\ \hline \texttt{bank} & Average & 0.045 & 0.09 & 0.0 & \textbf{0.5567}\\ \hline \texttt{bank} & Complete & 0.5 & 0.05 & 0.0 & \textbf{0.3327}\\ \hline \texttt{bank} & Single & 0.0005 & 1.0 & 0.0 & \textbf{0.8099}\\ \hline \end{tabular}} \end{center} \end{table}

2005.03197v4

\begin{table}[htbp] \caption{(MFC Results) FHAC, AFHAC-R, and AFHAC-V}\label{table3} \begin{center} %\begin{tabularx}{0.642\linewidth}{|p{2.6cm}|p{1.6cm}|p{2cm}|p{1cm}|p{1.6cm}|} \tabcolsep=0.10cm \resizebox{0.42\textwidth}{!}{ \begin{tabular}{|c|c|c|c|} \hline \textbf{Dataset} & \textbf{MFC (FHAC)} & \textbf{MFC (AFHAC-V)} & \textbf{MFC (AFHAC-R)}\\ \hline \hline \texttt{creditcard} & \textbf{1.0} & 2.8505 & 1.5825 \\ \hline \texttt{census} & \textbf{0.0853} & 4.0 & 4.0 \\ \hline \texttt{bank} & 0.6679 & \textbf{0.03648} & 2.0027\\ \hline \end{tabular}} \end{center} \end{table}

2005.03197v4

\begin{table}[htbp] \caption{(Balance Results) FHAC, AFHAC-R, AFHAC-V}\label{table4} \begin{center} %\begin{tabularx}{0.642\linewidth}{|p{2.6cm}|p{1.6cm}|p{2cm}|p{1cm}|p{1.6cm}|} \tabcolsep=0.09cm \resizebox{0.42\textwidth}{!}{ \begin{tabular}{|c|c|c|c|} \hline \textbf{Dataset} & \textbf{Balance(FHAC)} & \textbf{Balance(AFHAC-V)} & \textbf{Balance(AFHAC-R)}\\ \hline \hline \texttt{creditcard} & \textbf{0.427} & 0.0 & 0.416 \\ \hline \texttt{census} & \textbf{0.8865} & 0.2 & 0.0 \\ \hline \texttt{bank} & 0.5567 & \textbf{0.9474} & 0.0\\ \hline \end{tabular}} \end{center} \end{table}

2110.07519v1

\begin{table}[tb] \centering \caption{Index expansion rate (index size as a percentage of the original data size).}\label{table5} \hspace*{-0.3cm} \begin{tabular}{|c|c|c|c|} \hline &Synthetic&Seismic&SALD\\ & {100GB} & {100GB} & {100GB} \\ & {100M series} & {100M series} & {200M series} \\ \hline index expansion rate&5.7\%& 5.1\%& 10.5\%\\ \hline \end{tabular} \end{table}

2110.07519v1

\begin{table}[tb] % \centering % \scriptsize % \caption{Query answering algorithms comparison: number of times an operation is executed (average over 100 queries). }\label{table4} % \begin{tabular}{|c|c|c|c|c|c|} % \hline % & {\bf ParIS} & {\bf ParIS-TS} & {\bf ParIS-TS-LB} & {\bf MESSI-sq} & {\bf MESSI-mq} \\ % \hline % Insert node &n/a&69K& 69K& 15K& 15K\\ % \hline % Delete node&n/a&20K &20K& 11K& 11K\\ % \hline % LBD calculation &100M& 69K &9M& 9M &9M\\ % \hline % RD calculation &112K &9M &52K &54K &54K\\ % \hline % \end{tabular} %\end{table}

2110.07519v1

\begin{table}[tb] \centering \makeatletter\def\@captype{table}\makeatother \caption{Update Frequency of the BSF array (Euclidean distance).} { \begin{tabular}{|c|c|c|c|c|} \hline %Number of nearest neighbors & {\bf 1-NN} & {\bf 5-NN} & {\bf 10-NN} & {\bf 50-NN}\\ \hline number of BSF& & & & \\ %updates/query&11.97&20.92&45.58&258.04\\ updates/query&11.9&20.9&45.6&258.1\\ \hline BSF update time& & & &\\ %$\mu$sec/query&0.51&5.07&19.12&186.47\\ $\mu$sec/query&0.5&5.1&19.1&186.5\\ \hline % BSF update time\\ query time \% &0.001\% & 0.011\% & 0.038\% & 0.331\% BSF update time& & & &\\ query time \% &0.001\% & 0.01\% & 0.04\% & 0.3\% \\ \hline \end{tabular} } % font size \vspace*{0.1cm} \label{table2} \end{table}

2110.07519v1

\begin{table}[tb] \centering \makeatletter\def\@captype{table}\makeatother \caption{Update Frequency of the BSF array (DTW distance, 5\% warping).} { \begin{tabular}{|c|c|c|c|c|} \hline %Number of nearest neighbors & {\bf 1-NN} & {\bf 5-NN} & {\bf 10-NN} & {\bf 50-NN}\\ \hline number of BSF& & & & \\ %updates/query&22.71& 83.94& 160.2& 672.35\\ updates/query&22.7& 83.9& 160.2& 672.4\\ \hline BSF update time& & & &\\ %$\mu$sec/query&4.92&19.95& 50.21&473.27 $\mu$sec/query&4.9&19.9& 50.2&473.3 \\ \hline % BSF update time\\ query time \% &0.001\% & 0.011\% & 0.038\% & 0.331\% BSF update time& & & &\\ query time \textperthousand % &0.0072\textperthousand % &0.027\textperthousand % &0.062\textperthousand % &0.54\textperthousand &0.007\textperthousand &0.03\textperthousand &0.06\textperthousand &0.5\textperthousand \\ \hline \end{tabular} } % font size \vspace*{0.1cm} \label{tabledtw5} \end{table}

2110.07519v1

\begin{table}[tb] \centering \makeatletter\def\@captype{table}\makeatother \caption{Update Frequency of the BSF array (DTW distance, 10\% warping).} { \begin{tabular}{|c|c|c|c|c|} \hline %Number of nearest neighbors & {\bf 1-NN} & {\bf 5-NN} & {\bf 10-NN} & {\bf 50-NN}\\ \hline number of BSF& & & & \\ %updates/query&45.83& 124.73& 221.36& 854.04\\ updates/query&45.8& 124.7& 221.4& 854.1\\ \hline BSF update time& & & &\\ %$\mu$sec/query&11.91 &31.46&72.51&574.22 $\mu$sec/query&11.9 &31.5&72.5&574.2 \\ \hline % BSF update time\\ query time \% &0.001\% & 0.011\% & 0.038\% & 0.331\% BSF update time& & & &\\ query time \textperthousand % &0.0031\textperthousand& % 0.0076\textperthousand& % 0.0016\textperthousand& % 0.12\textperthousand &0.003\textperthousand& 0.008\textperthousand& 0.002\textperthousand& 0.1\textperthousand \\ \hline \end{tabular} } % font size \vspace*{0.1cm} \label{tabledtw10} \end{table}

1811.01315v2

\begin{table}[!h] \centering \caption{List of Abbreviations and Acronyms.} \begin{tabular}{c|c} \hline MNL & Multinomial logit \\ \hline NB & Naive Bayes \\ \hline CART & Classification and regression trees \\ \hline RF & Random forest \\\hline BOOST & Boosting trees \\\hline BAG & Bagging trees \\\hline SVM & Support vector machines \\\hline NN & Neural networks \\\hline AIC & Akaike information criterion \\\hline BIC & Bayesian information criterion \\\hline Min & Minimum \\\hline Max & Maximum \\\hline SD & Standard deviation \\\hline SP & Stated-preference \\\hline RP & Revealed-preference \\\hline IIA & Independence of irrelevant alternatives \\ \hline PT & Public transit \\ \hline \end{tabular} \label{tab:acy} \end{table}

1811.01315v2

\begin{table}[!t] \centering \caption{List of Symbols and Notations Used in the Paper} \footnotesize \begin{tabular}{c|c} \hline \bf{} Symbols & \bf{} Description \\ \hline $K$ & Total number of alternatives \\ \hline $N$ & Total number of observations \\ \hline $P$ & Total number of features\\ \hline $\boldsymbol{X}$ & Input data for logit models containing $P$ features with $N$ observations for $K$ alternatives \\ \hline $\boldsymbol X_{k,p}$ & Feature $p$ for alternative $k, k = 1, ..., K$ of $\boldsymbol X$\\ \hline $\boldsymbol{X}_{k,-p}$ & All the features except $p$ for alternative $k, k = 1, ..., K$ of $\boldsymbol X$\\ \hline $\boldsymbol{X}_{ik}$ & A row-vector for the $i$th observation for alternative $k, k = 1, ..., K$ \\ \hline $\boldsymbol{X}_k$ & Input data for alternative $k$, $\boldsymbol{X}_k = [\boldsymbol{X}_{.k1}; ...; \boldsymbol{X}_{.kP}]$ where $\boldsymbol{X}_{.kp} = [X_{1kp}, ..., X_{Nkp}]$\\ \hline $\boldsymbol{X}_i$ & The $i$th observation of $\boldsymbol{X}$, $\boldsymbol{X}_i = [\boldsymbol{X}_{i.1}, ..., \boldsymbol{X}_{i.P}]$ where $\boldsymbol{X}_{i.p} = [X_{i1p}; ...; X_{iKp}]$ \\ \hline $\boldsymbol{X}_p$ & The feature $p$ of $\boldsymbol{X}$, $\boldsymbol{X}_p = [\boldsymbol{X}_{1.p}, ..., \boldsymbol{X}_{N.p}]$ where $\boldsymbol{X}_{i.p} = [X_{i1p}; ...; X_{iKp}]$\\ \hline $\boldsymbol{Z}$ & Input data for machine-learning models containing $P$ features and $N$ observations\\ \hline $\boldsymbol Z_{p}$ & Feature $p$ of $\boldsymbol{Z}$\\ \hline $\boldsymbol Z_{-p}$ & All the features except $p$ of $\boldsymbol{Z}$\\ \hline $\boldsymbol{Z}_i$ & $i$th observation of $\boldsymbol{Z}, \boldsymbol{Z}_i = [Z_{i1}, ..., Z_{iP}$]\\ \hline $U_k(\boldsymbol{X}_k|\boldsymbol{\beta}_k)$ & Utility function for mode $k$ \\ \hline $\boldsymbol{\beta}_k$ & Parameter vector for alternative $k$ of MNL model\\ \hline $\boldsymbol{\beta}$ & Parameter matrix of MNL model, $\boldsymbol{\beta} = [\boldsymbol{\beta}_1,..., \boldsymbol{\beta}_K]$ \\ \hline $\hat{\boldsymbol{\beta}}$ & Estimated parameter matrix of MNL model \\ \hline $\boldsymbol{\varepsilon}_{k}$ & Random error for alternative $k$ of MNL model \\ \hline $\boldsymbol{Y}$ & Output mode choice data \\ \hline $\hat{Y}_i$ & Estimated mode choice for observation $i$ \\ \hline $\boldsymbol{\theta}$ & Parameter or hyperparameter vector for machine-learning models\\ \hline $\hat{\boldsymbol{\theta}}$ & Estimated parameter or hyperparameter vector\\ \hline $f(\boldsymbol{Z}|\boldsymbol{\theta})$ & Machine-learning models based on $\boldsymbol{Z}$ and $\boldsymbol{\theta}$ \\ \hline $p_{ik}$ & Probability of choosing alternative $k$ of observation $i$ \\ \hline $\hat{p}_{ik}$ & Predicted probability for choosing alternative $k$ of observation $i$\\ \hline $I_k(\hat{Y}_i)$ & Indicator function that equals to 1 if $\hat{Y}_i = k$ \\ \hline $P_k(\boldsymbol{X}|\hat{ \boldsymbol{\beta}})$ & Aggregate level prediction for mode $k$ based on $\boldsymbol{X}$ and $\hat{ \boldsymbol{\beta}}$ for logit models\\ \hline $Q_k(\boldsymbol{Z}|\hat{ \boldsymbol{\theta}})$ & Aggregate level prediction for mode $k$ based on $\boldsymbol{Z}$ and $\hat{ \boldsymbol{\theta}}$ for machine-learning models\\ \hline $E_k(\cdot)$ & Arc elasticity for alternative $k$\\ \hline $M_k(\cdot)$ & Marginal effect for alternative $k$\\ \hline $\Delta$ & Constant\\ \hline \end{tabular} \label{tab:symbol_des} \end{table}

1811.01315v2

\begin{table}[!] \caption{Comparison Between Logit and Machine-Learning Models} \footnotesize \resizebox{1\textwidth}{!}{% <------ Don't forget this % \begin{tabular}{p{3.2cm}|p{8.8cm}|p{7.5cm}} \hline \textbf{} & \textbf{Logit Models} & \textbf{Machine-Learning Models} \Tstrut\Bstrut \\ \hline \multirow{3}{*}{\textbf{Model formulation}} & $U_{k}(\boldsymbol{X}_k|\boldsymbol{\beta}_k) = \boldsymbol{\beta}_k^T \boldsymbol{X}_{k} + \boldsymbol{\varepsilon}_{k}$ & $\boldsymbol{Y} = f(\boldsymbol{Z}|\boldsymbol{\theta}), \boldsymbol{Y} \in \{1, …, K\}$ \Tstrut \\ & $p_{ik} = \frac{\exp {(\boldsymbol{\beta}}_k^T \boldsymbol{X}_{ik})}{\sum_{p=1}^K \exp {(\boldsymbol{\beta}}_p^T \boldsymbol{X}_{ik})}, k \in \{1, ..., K\}$ \Tstrut & \\ \hline \textbf{Commonly used model type} & MNL, mixed logit, nested MNL, generalized MNL & NB, CART, BAG, BOOST, RF, SVM, NN \Tstrut\Bstrut \\ \hline \textbf{Prediction type} & Class probability: $p_{i1}, …, p_{iK}$ & Classification: $k, k \in \{1, ..., K\}$ \Tstrut\Bstrut \\ \hline \textbf{Input data} & $\boldsymbol{X}$ & $\boldsymbol{Z}$ \\ \hline \textbf{Model topology} & Layer structure & Layer structure, tree structure, case-based reasoning, etc. \Tstrut\Bstrut \\ \hline \textbf{Optimization method} & Maximum likelihood estimation, simulated maximum likelihood & Back propagation, gradient descent, recursive partitioning, structural risk minimization, maximum likelihood, etc. \\ \hline \textbf{Evaluation criteria} & (Adjusted) McFadden's pseudo $R^2$, AIC, BIC & Resampling-based measures, e.g., cross validation \Tstrut\Bstrut \\ \hline \textbf{Individual-level mode prediction} & $\argmax_k (\hat{p}_{i1}, ..., \hat{p}_{iK})$ & $\hat{Y}_i$ \Tstrut\Bstrut \\ \hline \textbf{Aggregate-level mode share prediction} & $P_k(\boldsymbol{X}_k|\hat{ \boldsymbol{\beta}}_k) = \sum_i^N \hat{p}_{ik}/N$ & $Q_k(\boldsymbol{Z}|\hat{ \boldsymbol{\theta}}) = \sum_i^N \hat{p}_{ik}/N$ \Tstrut \\ \hline \textbf{Variable importance} & Standardized Beta coefficients & Variable importance, computed by using Gini index, out-of-bag error, and many others \Tstrut\Bstrut \\ \hline \textbf{Variable effects} & Sign and magnitude of Beta coefficients & Partial dependence plots \Tstrut\Bstrut \\ \hline \textbf{Arc elasticity of feature $p$ for alternative $k$} & $E_k(\boldsymbol{X}_{k,p}) = \frac{[P_k(\boldsymbol{X}_{k,-p}, \boldsymbol X_{k,p} \cdot (1+\Delta) | \hat{ \boldsymbol{\beta}}_k) - P_k(\boldsymbol{X}_{k} | \hat{ \boldsymbol{\beta}}_k)]/P_k(\boldsymbol{X}_{k} | \hat{ \boldsymbol{\beta}}_k)}{|\Delta|},$ $k \in \{1, ..., K\}$ \Tstrut & $E_k(\boldsymbol{Z}_{p}) = \frac{[Q_k(\boldsymbol{Z}_{-p}, \boldsymbol Z_{p} \cdot (1+\Delta) | \hat{ \boldsymbol{\theta}}_k) - Q_k(\boldsymbol{Z} | \hat{ \boldsymbol{\theta}}_k)]/Q_k(\boldsymbol{Z} | \hat{ \boldsymbol{\theta}}_k)}{|\Delta|},$ $k \in \{1, ..., K\}$ \Tstrut \\ \hline \textbf{Marginal effects of feature $p$ for alternative $k$} & $M_k(\boldsymbol{X}_{k,p}) = \frac{P_k(\boldsymbol{X}_{k,-p}, \boldsymbol{X}_{k,p} +\Delta) | \hat{ \boldsymbol{\beta}}_k) - P_k(\boldsymbol{X}_{k} | \hat{ \boldsymbol{\beta}}_k)}{|\Delta|},$ $k \in \{1, ..., K\}$ \Tstrut & $M_k(\boldsymbol{Z}_{p}) = \frac{Q_k(\boldsymbol{Z}_{-p}, \boldsymbol Z_{p} + \Delta) | \hat{ \boldsymbol{\theta}}_k) - Q_k(\boldsymbol{Z} | \hat{ \boldsymbol{\theta}}_k)}{|\Delta|},$ $k \in \{1, ..., K\}$ \Tstrut \\ \hline \end{tabular} } \label{tab:comparison} \end{table}

1811.01315v2

\begin{table}[!t] \caption{Ranking of Variable Importance for RF, NN, MNL, and Mixed Logit} \begin{tabular}{c|cccc} \hline \textbf{Variable} & \textbf{RF} & \textbf{NN} & \textbf{MNL} & \textbf{Mixed logit} \\ \hline TT\_Walk & 1 & 16 & 2 & 2 \\ TT\_Drive & 2 & 14 & 4 & 5 \\ TT\_Bike & 3 & 13 & 1 & 1 \\ TT\_PT & 4 & 11 & 3 & 3 \\ Current\_Mode\_Bike & 5 & 2 & 5 & 10 \\ PT\_Access & 6 & 8 & 13 & 16 \\ Bike\_Walkability & 7 & 6 & 16 & 13 \\ Income & 8 & 10 & 14 & 15 \\ CarPerCap & 9 & 7 & 11 & 14 \\ Current\_Mode\_Walk & 10 & 1 & 6 & 12 \\ Rideshare & 11 & 9 & 9 & 9 \\ Transfer & 12 & 5 & 8 & 8 \\ Wait\_Time & 13 & 15 & 10 & 11 \\ Female & 14 & 3 & 15 & 17 \\ Parking\_Cost & 15 & 12 & 12 & 4 \\ Current\_Mode\_Car & 16 & 4 & 7 & 6 \\ Current\_Mode\_PT & / & / & 17 & 7 \\ \hline \end{tabular} \label{tab:VarImp} \end{table}

1612.03079v2

\begin{table}[t] \centering \small % \subfloat[Benchmark Datasets]{ \begin{tabular}[b]{ | l | l | l | c | c | c | } \hline \textbf{Dataset} & \textbf{Type} & \textbf{Size} & \textbf{Features} & \textbf{Labels} \\ \hline MNIST \cite{mnist} & Image & 70K & 28x28 & 10 \\ CIFAR \cite{cifardata} & Image & 60k & 32x32x3 & 10 \\ ImageNet \cite{imagenet} & Image & ~1.26M & 299x299x3 & 1000 \\ Speech \cite{timit} & Sound & 6300 & 5 sec. & 39 \\ \hline \end{tabular} \vspace{-4mm} \caption{\small \textbf{Datasets.} The collection of real-world benchmark datasets used in the experiments.} \vspace{-5mm} \label{tab:datasets} \end{table}

1109.6846v2

\begin{table} \label{table:sham} \begin{center} \begin{tabular}{|c||c|c|} \hline observed degree & \# predicted ($E[N_{\delta,\rho^*}]$) & \# actual ($N_{\delta,\rho^*}$) \\ \hline $d_i\geq \delta =1$ & 8531 & 8492 \\ \hline $d_i\geq \delta =2$ & 1697 & 1635 \\ \hline $d_i\geq \delta =3$ & 234 & 229 \\ \hline $d_i\geq \delta =4$ & 24 & 28 \\ \hline $d_i\geq \delta =5$ & 2 & 4 \\ \hline \end{tabular} \caption{Fidelity of the predicted (mean) number of false positives and the observed number of false positives in the realization of the sham NKI dataset experiment shown in Fig. \ref{fig:pv1}}. \end{center} \end{table}

2305.02029v1

\begin{table}[!htbp] \begin{tabular}{|p{0.88\textwidth}|p{0.15\textwidth}|} \hline High Probability Words & Suggested topic\\ \hline 0.018*"quote" + 0.013*"ebay" + 0.010*"finance" + 0.009*"premium" + 0.009*"level" + 0.008*"performance" + 0.007*"new\_car" + 0.007*"expensive" + 0.007*"struggle" + 0.006*"advance" & package\\ \hline 0.019*"data" + 0.015*"flag" + 0.013*"meet" + 0.011*"retail\_check" + 0.010*"price\_indicator" + 0.010*"spec" + 0.008*"sit" + 0.008*"valuations" + 0.007*"group" + 0.007*"price\_flags" & price indicator flags\\ \hline 0.012*"request" + 0.012*"admin\_fees" + 0.011*"video" + 0.007*"find" + 0.007*"image" + 0.006*"actually" + 0.006*"query" + 0.006*"frustrate" + 0.005*"spec" + 0.005*"unhappy" & unhappy\\ \hline 0.017*"image" + 0.013*"rat" + 0.012*"new\_car" + 0.010*"highly" + 0.009*"upload" + 0.008*"reply" + 0.008*"award" + 0.007*"consumers" + 0.006*"info" + 0.006*"message" & live chat\\ \hline 0.012*"text" + 0.011*"valuations" + 0.011*"product" + 0.010*"chat" + 0.009*"lose" + 0.007*"tech" + 0.007*"margin" + 0.006*"platform" + 0.006*"retail" + 0.006*"higher" & valuations\\ \hline 0.010*"retract" + 0.008*"close" + 0.008*"open" + 0.007*"watch" + 0.007*"webinar" + 0.006*"book" + 0.006*"phone" + 0.006*"process" + 0.006*"answer" + 0.006*"charge" & process related\\ \hline 0.011*"staff" + 0.010*"coronavirus" + 0.010*"reduce" + 0.010*"lockdown" + 0.009*"canx" + 0.009*"plan" + 0.008*"struggle" + 0.008*"online" + 0.008*"june" + 0.008*"continue" & coronavirus\\ \hline 0.016*"lockdown" + 0.010*"june" + 0.010*"collect" + 0.008*"open" + 0.008*"retract" + 0.007*"confuse" + 0.007*"follow" + 0.007*"aware" + 0.007*"extend" + 0.007*"appreciate" & lockdown extensions\\ \hline 0.061*"xxxemailxxx" + 0.023*"subject" + 0.015*"group" + 0.013*"kind" + 0.009*"xxxtelephonexxx" + 0.008*"sit" + 0.008*"retail" + 0.007*"lead" + 0.007*"manheim" + 0.006*"option" & no recognised subject\\ \hline 0.027*"year" + 0.016*"experian" + 0.008*"car\_gurus" + 0.007*"ebay" + 0.006*"meet" + 0.006*"zuto" + 0.006*"july" + 0.005*"achieve" + 0.005*"award" + 0.005*"normal" & rival valuation products\\ \hline \end{tabular} \caption{Topic modelling results from the AutoTrader note corpus, with sector expert led topic naming suggestions.} \label{table:1} \end{table}

2305.02029v1

\begin{table}[h] \centering \begin{tabular}{|c|c|} \hline Clean & Baseline\\ \hline Valuation &0.97 \\ \hline Price &0.76 \\ \hline Package &0.78 \\ \hline Cancellation &0.63 \\ \hline Stock &0.68 \\ \hline Tech &0.86 \\ \hline Billing &0.56 \\ \hline \end{tabular} \caption{ Baseline NDCG evaluations for the clean data. } \label{table:2} \end{table}

2305.02029v1

\begin{table}[h] \centering \begin{tabular}{|c|c|} \hline Pre-processed & Baseline\\ \hline Valuation &0.97 \\ \hline Price &0.76 \\ \hline Package &0.78 \\ \hline Cancellation &0.63 \\ \hline Stock &0.68 \\ \hline Tech &0.86 \\ \hline Billing &0.57 \\ \hline \end{tabular} \caption{ Baseline NDCG evaluations for the pre-processed data. } \label{table:3} \end{table}

2305.02029v1

\begin{table}[h!] \centering \begin{tabular}{|c|c|c|} \hline Topic & Score & Difference from baseline\\ \hline Valuation &0.96 & -0.01\\ \hline Price &0.72 & -0.04\\ \hline Package &0.74 & -0.04\\ \hline Cancellation &0.60 & -0.03\\ \hline Stock &0.67 & -0.01\\ \hline Tech &0.92 & +0.06\\ \hline Billing &0.55 & -0.02\\ \hline \end{tabular} \caption{NDCG evaluations for the query ``tech issue" on the pre-processed data.} \label{table:4} \end{table}

2305.02029v1

\begin{table}[h!] \centering \begin{tabular}{|c|c|c|} \hline Topic & Score & Difference from baseline\\ \hline Valuation &0.97 & 0.00\\ \hline Price &0.86 & +0.10\\ \hline Package &0.79 & +0.01\\ \hline Cancellation &0.66 & +0.03\\ \hline Stock &0.68 & 0.00\\ \hline Tech &0.82 & -0.04\\ \hline Billing &0.54 & -0.03\\ \hline \end{tabular} \caption{NDCG evaluations for the query ``too expensive" on the pre-processed data.} \label{table:5} \end{table}

2305.02029v1

\begin{table}[h] \centering \begin{tabular}{|c|c|c|} \hline Topic & Score & Difference from baseline\\ \hline Valuation &0.96 & -0.01\\ \hline Price &0.75 & -0.01\\ \hline Package &0.74 & -0.04\\ \hline Cancellation &0.65 & -0.02\\ \hline Stock &0.64 & -0.04\\ \hline Tech &0.84 & -0.02\\ \hline Billing &0.68 & +0.11\\ \hline \end{tabular} \caption{NDCG evaluations for the query ``send money" on the pre-processed data.} \label{table:6} \end{table}

1807.08372v1

\begin{table}[h!] \begin{smallermathTable} \scriptsize{ \centering \begin{tabular}[t]{p{1.52cm}<{\centering}|p{1.3cm}<{\centering}|p{1.3cm}<{\centering}|p{1.3cm}<{\centering}|p{0.98cm}<{\centering}} \hline TBox Axi.: 541 & Concept Ast. Ent.: 1824 &Role Ast. Ent.: 4528 & Individual: 1159 & Ext. Axi.: $\sim21$ K \\\hline \hline Parameters ($\sigma, \kappa, \tau$) & Root Concept Ast. Ent.& Root Role Ast. Ent. & Root Individual & External Axioms\\\hline P1:($.90, 1, .40$) & $1105$ ($61\%$) & $3805$ ($84\%$) & $1103$ ($95\%$) & $\sim$ $20$ K \\\hline P2:($.93, 1, .43$) & $990$ ($54\%$) & $3459$ ($76\%$) & $1080$ ($93\%$) & $\sim$ $19$ K \\\hline P3:($.96, 1, .46$) & $540$ ($30\%$) & $1816$ ($40\%$)& $872$ ($75\%$) & $\sim$ $16$ K \\\hline P4:($.99, 1, .49$) & $305$ ($17\%$) &$980$ ($22\%$) &$510$ ($44\%$) & $6271$ \\\hline P5:($.99, 2, .49$) & $157$ ($8.6\%$) & $402$ ($8.9\%$)& $108$ ($9.3\%$) & $615$ \\\hline \end{tabular} \vspace{-0.25cm} \caption{\label{res:count} Average Number of Root Entailments, Root Individuals and External Axioms per Learning Domain.} } \end{smallermathTable} \end{table}

2406.18112v1

\begin{table} \caption{Averaged compute time of time steps in milliseconds for different parts of the in transit and hybrid analysis.}\label{tab-benchmark} \centering \begin{tabular}{|l|l|l|l|l|} \hline & \multicolumn{2}{c|}{\emph{Slice} Pipeline} & \multicolumn{2}{c|}{\emph{Resampling} Pipeline} \\ \cline{2-5} & In Transit & Hybrid & In Transit & Hybrid \\ \hline Simulation Time (ms) & 15860 & 15835 & 33539 & 31677\\ \hline Reduction Time (ms) & 0 & 319 & 0 & 2905\\ \hline Data Transfer Time (ms)& 3415 & 6.56 & 10957 & 48.7 \\ \hline \hline Total Time (ms)& 19275 & 16161 & 44497 & 34632 \\ \hline Total Gain & \multicolumn{2}{c|}{16.16\%} & \multicolumn{2}{c|}{22.17\%}\\ \hline \end{tabular} \end{table}

1710.09593v2

\begin{table}[!htb] \centering \caption{The characteristics of the used Machines} \label{capa} \begin{tabular}{|c|c|c|c|} \hline \textbf{Machine's name} & \textbf{Operating System} & \textbf{Processor} & \textbf{Memory} \\\hline Dell-XPS L421X& \pbox{50cm}{Ubuntu \\ (V.14.04 LTS) }& \pbox{50cm}{1.8GHz*4 \\ Intel Core i5} & 8 GB \\ \hline Dell-Inspiron-3721 & \pbox{50cm}{Ubuntu \\ (V.14.04 LTS)} & \pbox{50cm}{2.00GHz*4 \\Intel Core i5 }& 4 GB \\\hline Dell-Inspiron-3521 & \pbox{50cm}{ Ubuntu \\(V.16.04 LTS)} &\pbox{50cm}{ 1.8 GHz*4 \\Intel Core i5 } & 6 GB \\\hline iMac-Early 2010 & \pbox{50cm}{cinux Mint \\ (V.17.1 Rebecca) }& \pbox{50cm}{3.06GHz*2} & 4 GB \\ \hline Dell-Inspiron-5559 & \pbox{50cm}{Ubuntu \\ (V.16.04 LTS) }& \pbox{50cm}{2.30GHz*4 \\ Intel Core i5} & 8 GB \\\hline iMac-Early 2009 & \pbox{50cm}{OS X El Capitan \\(V.10.11.6)} &\pbox{50cm}{ 2.93 *2 GHz \\Intel Core Due}& 8 GB \\\hline MacBook Air & \pbox{50cm}{OS X El Capitan \\(V.10.11.3)} &\pbox{50cm}{ 1.6 *2 GHz \\Intel Core i5 }& 8 GB \\\hline \end{tabular} \end{table}

1710.09593v2

\begin{table}[!htb] \centering \caption{Datasets} \label{DS} \begin{tabular}{|c|c|c|} \hline \textbf{Benchmark}& \textbf{Size} & \textbf{Descriptions} \\\hline \textbf{D1}& 10,000 Points & \begin{tabular}[c]{@{}c@{}}Different shapes, with \\ some clusters surrounded\\ by others\end{tabular} \\\hline \textbf{D2}& 30,000 Points& \begin{tabular}[c]{@{}c@{}}2 small circles,\\ 1 big circle\\ and 2 linked ovals\end{tabular} \\ \hline \end{tabular} \end{table}

2306.00454v1

\begin{table} \centering % \begin{tabular}{l|cc} \hline & Reaction rate for $d(p,\gamma)^{3}{\rm He}$ & $D_{p}$ data \tabularnewline \hline \hline { } Case I & Adelberger et al. & no-data\tabularnewline { } Case II & Adelberger et al. & Cooke et al. + offset \tabularnewline { } Case III & Adelberger et al. & Cooke et al.\tabularnewline { } Case IV & Marcucci et al. & Cooke et al.\tabularnewline { } Case V & Marcucci et al. & recent 11 weighted mean\tabularnewline { } Case VI & Pisanti et al. & Cooke et al.\tabularnewline { } Case VII & Pisanti et al. & recent 11 weighted mean\tabularnewline \hline \hline \end{tabular}\caption{\label{tab:Dp_treatment} Combinations of the data of $D_{p}$ and theoretical treatment of the reaction rate for $d(p,\gamma)^{3}{\rm He}$ used in our analysis. } \end{table}

2103.08664v1

\begin{table}[t] \centering \caption{Test accuracy of different training strategies on Physionet data: left fist VS right fist (task 2); both fists VS both feet (task 4).} \begin{tabular}{|c|cc|}\hline motor imagery task & task 2 & task 4 \\ \hline conventional learning & 59.6\% & 66.6\% \\ transfer learning & 56.0\% & 62.4\% \\ meta-learning & \textbf{64.5\%} & \textbf{68.2\%} \\ \hline \end{tabular} \label{tab:5} \end{table}

2103.08664v1

\begin{table}[b] \centering \caption{Test accuracy on Physionet data after online sample filtering} \begin{tabular}{|c|cc|} \hline motor imagery task & task 2 & task 4 \\ \hline without meta-learning & 67.9\% & 68.8\% \\ with meta-learning & \textbf{80.6\%} & \textbf{79.7\%}\\ \hline \end{tabular} \label{tab:7} \end{table}

2109.05142v1

\begin{table}[ht] \centering \begin{tabular}{|c|c|p{1.3in}|} \hline \textbf{Data Source}& \textbf{Data Model} &\multicolumn{1}{c|}{\textbf{Polystore Placement}} \\ \hline Patents &Relational &PostgreSQL, Text in Solr \\ \hline News articles &Structured Text &Solr, Entity Network in Neo4J \\ \hline Federal Spending&Relational &PostgreSQL \\ \hline Company Networks&Graph &Neo4J \\ \hline \end{tabular} \vspace{1em} \caption{Data from a source are processed and placed into different stores.} \label{tab:sources} \vspace{-1.3em} \end{table}

1911.00108v2

\begin{table}[H] \centering \small \begin{tabular}{|l|l|l|} \hline \textbf{Method} & \textbf{Average Accuracy} & \textbf{stdev} \\ \hline TPOT & 0.816 & 0.159 \\ \hline auto-sklearn(V) & 0.796 & 0.168 \\ \hline auto-sklearn(E) & 0.805 & 0.165 \\ \hline RankML \#1 rank & 0.786 & 0.169 \\ \hline RankML Max top-5 rank &0.819 & 0.154 \\ \hline RankML Max top-10 rank &\textbf{0.827} & 0.152 \\ \hline \end{tabular} \caption[Average accuracy Results]{Average accuracy results across 149 classification datasets.} \label{table:1} \end{table}

1911.00108v2

\begin{table}[H] \resizebox{\columnwidth}{!}{% \begin{tabular}{|l|c|c|c|} \hline \multicolumn{1}{|c|}{\multirow{2}{*}{\textbf{Method}}} & \multicolumn{3}{c|}{\textbf{Number of Datasets with BOC Performance(\%)}} \\ \cline{2-4} \multicolumn{1}{|c|}{} & \textbf{TPOT} & \textbf{auto-sklearn(V)} & \textbf{auto-sklearn(E)} \\ \hline RankML \#1 rank & 65(49\%) & 74(55\%) & 68(51\%) \\ \hline RankML Max top-5 rank & 102(73\%) & 112(80\%) & 112(80\%) \\ \hline RankML Max top-10 rank & \textbf{110(79\%)} & \textbf{119(85\%)} & \textbf{118(84\%)} \\ \hline \end{tabular}% } \caption{The number of classification datasets each method got better or comparable(BOC) results against baselines (percentage is out of valid datasets).} \label{table:2} \end{table}

1911.00108v2

\begin{table}[H] \centering \resizebox{\columnwidth}{!}{% \begin{tabular}{|m{3.3cm}|m{2cm}|m{1.1cm}|} \hline \multicolumn{1}{|c|}{\textbf{Primitive}} & \multicolumn{1}{c|}{\textbf{Family}} & \multicolumn{1}{c|}{\textbf{\begin{tabular}[c]{@{}c@{}}Avg \% \\ of appearances\end{tabular}}} \\ \hline MaxAbsScaler & Data pre-processing & 16\% \\ \hline StandardScaler & Data pre-processing & 11\% \\ \hline KNeighborsClassifier & Predictive models & 10\% \\ \hline RandomForestClassifier & Predictive models& 4\% \\ \hline PCA & Feature pre-processing & 0.4\% \\ \hline \end{tabular}% } \caption{A selection of primitives used in RankML recommended pipelines for classification. "Avg \% of appearances" is out of all primitives. } \label{table:4} \end{table}

1911.00108v2

\begin{table} \resizebox{\columnwidth}{!}{% \begin{tabular}{|l|c|c|c|} \hline \multicolumn{1}{|c|}{\multirow{2}{*}{\textbf{Method}}} & \multicolumn{3}{c|}{\textbf{Number of Datasets with BOC Performance(\%)}} \\ \cline{2-4} \multicolumn{1}{|c|}{} & \textbf{TPOT} & \textbf{auto-sklearn(V)} & \textbf{auto-sklearn(E)} \\ \hline RankML \#1 rank & 30(39\%) & 35(46\%) & 26(34\%) \\ \hline RankML Max top-5 rank & 39(49\%) & 47(59\%) & 41(52\%) \\ \hline RankML Max top-10 rank & \textbf{45(57\%)} & \textbf{53(67\%)} & \textbf{45(57\%)} \\ \hline \end{tabular}% } \caption{The number of regression datasets each version of our approach got better or comparable results against the baselines (percentage is out of valid datasets).} \label{table:1_reg} \end{table}

1509.03329v1

\begin{table}[!ht] \caption{Summary of cuts applied in the \xeff\ analysis depending on the brightness of the source.} \begin{center} \begin{tabular}{|c|ccc|} \hline % \rowcolor{gray!10} Brightness &\phi12 & $\eta$ & \xeff\ \\% cut \\ % \rowcolor{gray!10} &\diffi&& \\ \hline \hline Very Bright & $>10$ & 0.1 & $<0.3$ \\ % \hline Bright & 1 ... 10 & 0.2 & $<0.3$ \\ % \hline Medium & 0.5 ... 1 & 0.4 & $<0.3$\\ % \hline Faint & $< 0.5 $ & 0.7 & $<0.3$ \\ % \hline % Very faint & $< 0.1 $ & 0.8 & 0.3 \\ % $<1 \% $ Crab &&& \\ % \hline \hline \end{tabular} \end{center} \label{tabcuts} \end{table}

1904.08994v1

\begin{table}[h!] \centering \begin{tabular}{c|l|l} \hline \textbf{Symbol} & \textbf{Meaning} & \textbf{Notes}\\ \hline $p_{z}$ & Data distribution over noise input $z$ & Usually, just uniform. \\ $p_{g}$ & The generator's distribution over data $x$ & \\ $p_{r}$ & Data distribution over real sample $x$ & \\ \hline \end{tabular} \end{table}

1904.08994v1

\begin{table}[h!] \centering \begin{tabular}{c|p{10cm}} \hline \textbf{Term} & \textbf{Explanation} \\ \hline Manifold & A topological space that locally resembles Euclidean space near each point. Precisely, when this Euclidean space is of dimension $n$, the manifold is referred as $n$-manifold. \\ Support & A real-valued function $f$ is the subset of the domain containing those elements which are not mapped to zero.\\ \hline \end{tabular} \end{table}

1907.09456v3

\begin{table}[h] \centering \caption{Summary of \texttt{SCSF} Tuning Parameters} \begin{tabular}{|c|l|c|} \hline \textbf{Param.} & \textbf{Description} & \textbf{Value} \\ \hline $k$ & Rank of the matrix factorization & $6$ \\ $\tau$ & Approximate quantile of the data to be fit & $0.85$ \\ $\mu_L$ & Weight of the smoothing term on the left matrix & $500$ \\ $\mu_R$ & Weight of the smoothing term on the right matrix & $1000$ \\ \hline \end{tabular} \label{tab:tuning-param} \end{table}

1907.09456v3

\begin{table}[b] \centering \caption{\texttt{SCSF} Tuning Parameter Grid Search} \begin{tabular}{|c|c|c|c|} \hline \textbf{Param.} & \textbf{Low Val.} & \textbf{Mid Val.} & \textbf{High Val.} \\ \hline $k$ & $4$ & $6$ & $8$ \\ $\tau$ & $.8$ & $.85$ & $.9$ \\ $\mu_L$ & $1e2$ & $5e2$ & $1e3$ \\ $\mu_R$ & $5e2$ & $1e3$ & $5e3$ \\ \hline \end{tabular} \label{tab:tuning-study} \end{table}

1907.09456v3

\begin{table}[b] \centering \caption{Summary of site selection for both methods} \begin{tabular}{|c|c|c|c|} \hline & & & \textbf{Unique} \\ & \textbf{Included Sites}& \textbf{Excluded Sites} & \textbf{Excluded Sites} \\ \hline \texttt{SCSF}& $551$ & $22$ & $19$ \\ \texttt{RdTools} & $387$ & $186$ & $183$\\ \hline \end{tabular} \label{tab:fleet-join} \end{table}

2211.08064v2

\begin{table}[h] \begin{tabular}{c|c} \hline Notations & Description \\ \hline $u$ & state variables of the physical system \\ \hline $\boldsymbol{x}$ & spatial or spatial-temporal coordinates \\ \hline $x$ & spatial coordinates \\ \hline $t$ & temporal coordinates \\ \hline $\theta$ & parameters for a physical system \\ \hline $w$ & weights of neural networks \\ \hline $\frac{\partial}{\partial x_i}$ & partial derivatives operator \\ \hline $\mathcal{D}^k_i$ & $\frac{\partial^k}{\partial x_i^k}$, $k$-order derivatives for variable $x_i$ \\ \hline $\nabla$ & nabla operator (gradient) \\ \hline $\Delta$ & Laplace operator \\ \hline $\int$ & integral operator \\ \hline $\mathcal{F}$ & differential operator representing the PDEs/ODEs \\ \hline $\mathcal{I}$ & initial conditions (operator) \\ \hline $\mathcal{B}$ & boundary conditions (operator) \\ \hline $\Omega$ & spatial or spatial-temporal domain of the system \\ \hline $\Theta$ & space of the parameters $\theta$ \\ \hline $W$ & space of weights of neural networks \\ \hline $\mathcal{L}$ & loss functions \\ \hline $\mathcal{L}_r$ & residual loss \\ \hline $\mathcal{L}_b$ & boundary condition loss \\ \hline $\mathcal{L}_i$ & initial condition loss \\ \hline $l_k$ & residual (error) terms \\ \hline $\| \cdummy \|$ & norm of a vector or a function \\ \hline \end{tabular} \caption{A table of mathematical notations.} \label{tb1} \end{table}

1903.07639v2

\begin{table}[ht!] \caption{\textbf{Elements of a data analysis.} This table describes eight elements that are used by the data analyst to build the data analysis.} \label{table-elements} \begin{tabular}{p{3cm}|p{11cm}} \hline \textbf{Element} & \textbf{Description} \\ \hline Narrative text & Expository phrases or sentences that describe what is happening in the data analysis in a human readable format \\ \hline Code & A series of programmatic instructions to execute a particular programming or scripting language \\ \hline Code comment & Non-executable code or text near or inline with code that describes the expected action/result of the surrounding code or provides context \\ \hline Data visualization & A plot, figure or graph illustrating a visual representation of the data. \\ \hline Narrative diagram & A diagram or flowchart without data \\ \hline Summary statistics & Numerical quantities derived from the data, such as the mean, standard deviation, etc. \\ \hline Table & An ordered arrangement of data or summaries of data in rows and columns \\ \hline Statistical model or computational algorithm & Mathematical model or algorithm concerning the underlying data phenomena or data-generation process, predictive ability, or computational algorithm \\ \hline \end{tabular} \end{table}

2103.03739v1

\begin{table}[t!] \centering \caption{Threat prioritization depending on likelihood and impact.} \label{tab:lin:threat-table:priorities} \scalebox{0.92}{ \begin{tabular}{|c|c|c|} \hline Likelihood & Impact & Priority \\ \hline \hline low & low & \multirow{3}{*}{low} \\ low & medium & \\ medium & low & \\ \hline \end{tabular} % \quad % \begin{tabular}{|c|c|c|} \hline Likelihood & Impact & Priority \\ \hline \hline low & high & \multirow{3}{*}{medium}\\ medium & medium & \\ high & low & \\ \hline \end{tabular} % \quad % \begin{tabular}{|c|c|c|} \hline Likelihood & Impact & Priority \\ \hline \hline medium & high & \multirow{3}{*}{high} \\ high & medium & \\ high & high & \\ \hline \end{tabular} } \end{table}

2103.03739v1

\begin{table}[t] \centering \caption{Overview of threat prioritization. Threats that are not effective due to our assumptions are not included in the table.} \label{tab:lin:threat-table:prioritization} \begin{tabular}{ | m{.6\textwidth}<{\raggedleft} || c | c || c | } \hline \multicolumn{1}{|c||}{Threat} & Likelihood & Impact & Priority \\ \hline \hline Linkability in one or more storages & medium & medium & medium \\ \hline Identifiability in one or more storages & low & high & medium \\ \hline Detectability of data existence & medium & low & low \\ \hline Detectability in communication between different trust domains & low & low & low \\ \hline Linkability of IP addresses in communication between different trust domains & low & medium & low \\ \hline Linkability of IP addresses in communication between different trust domains leads to identifiability & low & high & medium\\ \hline Non-repudiation of encrypted data & low & low & low \\ \hline Non-repudiation of communication between different trust domains & low & low & low \\ \hline Unawareness of the data owner & low & high & medium \\ \hline Non deletion of data in cloud storage & low & low & low \\ \hline \end{tabular} \end{table}

2003.09758v1

\begin{table}[htb!] \small \centering \begin{tabular}{|c|c|c|c|c|} \hline \rowcolor[HTML]{EFEFEF} \textbf{Dataset} & \textbf{\begin{tabular}[c]{@{}c@{}}Score \\ change\end{tabular}} & \textbf{\begin{tabular}[c]{@{}c@{}}Speed \\ (x faster)\end{tabular}} & \textbf{\begin{tabular}[c]{@{}c@{}}Tables \\ removed\end{tabular}} & \textbf{$\tau$} \\ \hline Taxi & -0.04\% & 3.18 & 10 & 24 \\ \hline Pickup & -15.35\% & 3.50 & 17 & 17 \\ \hline Poverty & -1.19\% & 5.87 & 36 & 15 \\ \hline School (S) & -1\% & 1.14 & 2 & 15 \\ \hline School (L) & -5\% & 1.32 & 39 & 17 \\ \hline \end{tabular} \caption{Performance of ARDA with \textsf{RIFS} and Tuple Rule as a table filtering step for real world datasets. Hyperparameter $\tau$ was optimized for each dataset.} \label{tab:TR_RIFS_prefilter} \end{table}

1305.2505v1

\begin{table}[t] \centering \begin{tabular}{|c|c|} \hline Hypothesis class & Rademacher Complexity\\\hline $\B_q(\norm{\W}_q)$ & $2\norm{\X}_p\norm{\W}_q\sqrt{\frac{p - 1}{n}}$\\\hline $\B_1(\norm{\W}_1)$& $2\norm{\X}_\infty \norm{\W}_1\sqrt{\frac{e\log d}{n}}$\\\hline \end{tabular} \caption{Rademacher complexity bounds for AUC maximization. We have $1/p+1/q = 1$ and $q > 1$.} \label{tab:rad-bounds-auc} \end{table}