Machine Learning

Authors and titles for March 2023

[176] arXiv:2303.03181 (cross-list from cs.LG) [pdf, other]

Title: MetaPhysiCa: OOD Robustness in Physics-informed Machine Learning

S Chandra Mouli, Muhammad Ashraful Alam, Bruno Ribeiro

Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)

[177] arXiv:2303.03187 (cross-list from cs.LG) [pdf, other]

Title: Boosting Differentiable Causal Discovery via Adaptive Sample Reweighting

An Zhang, Fangfu Liu, Wenchang Ma, Zhibo Cai, Xiang Wang, Tat-seng Chua

Comments: In proceedings of ICLR 2023

Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)

[178] arXiv:2303.03254 (cross-list from math.OC) [pdf, other]

Title: An Online Algorithm for Chance Constrained Resource Allocation

Yuwei Chen, Zengde Deng, Yinzhi Zhou, Zaiyi Chen, Yujie Chen, Haoyuan Hu

Comments: 5 pages, 5 figures. Accepted to ICASSP 2023. arXiv admin note: substantial text overlap with arXiv:2203.16818

Subjects: Optimization and Control (math.OC); Machine Learning (cs.LG); Machine Learning (stat.ML)

[179] arXiv:2303.03272 (cross-list from cs.LG) [pdf, other]

Title: Accelerated Rates between Stochastic and Adversarial Online Convex Optimization

Sarah Sachs, Hedi Hadiji, Tim van Erven, Cristobal Guzman

Comments: Extended version of 'Between Stochastic and Adversarial Online Convex Optimization: Improved Regret Bounds via Smoothness' by the same authors. arXiv admin note: text overlap with arXiv:2202.07554

Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)

[180] arXiv:2303.03293 (cross-list from cs.LG) [pdf, other]

Title: On Hierarchical Multi-Resolution Graph Generative Models

Mahdi Karami, Jun Luo

Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)

[181] arXiv:2303.03327 (cross-list from cs.LG) [pdf, other]

Title: Tight Bounds for $γ$-Regret via the Decision-Estimation Coefficient

Margalit Glasgow, Alexander Rakhlin

Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)

[182] arXiv:2303.03348 (cross-list from cs.LG) [pdf, other]

Title: Thompson Sampling for Linear Bandit Problems with Normal-Gamma Priors

Björn Lindenberg, Karl-Olof Lindahl

Comments: 27 pages, 2 figures

Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)

[183] arXiv:2303.03363 (cross-list from q-bio.BM) [pdf, other]

Title: Enhancing Activity Prediction Models in Drug Discovery with the Ability to Understand Human Language

Philipp Seidl, Andreu Vall, Sepp Hochreiter, Günter Klambauer

Comments: ICML version, 15 pages + 18 pages appendix

Subjects: Biomolecules (q-bio.BM); Computation and Language (cs.CL); Machine Learning (cs.LG); Machine Learning (stat.ML)

[184] arXiv:2303.03374 (cross-list from cs.LG) [pdf, other]

Title: To Stay or Not to Stay in the Pre-train Basin: Insights on Ensembling in Transfer Learning

Ildus Sadrtdinov, Dmitrii Pozdeev, Dmitry Vetrov, Ekaterina Lobacheva

Comments: Published in NeurIPS 2023. First two authors contributed equally

Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)

[185] arXiv:2303.03382 (cross-list from cs.LG) [pdf, other]

Title: Globally Optimal Training of Neural Networks with Threshold Activation Functions

Tolga Ergen, Halil Ibrahim Gulluk, Jonathan Lacotte, Mert Pilanci

Comments: Accepted to ICLR 2023

Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)

[186] arXiv:2303.03384 (cross-list from cs.LG) [pdf, other]

Title: Restoration-Degradation Beyond Linear Diffusions: A Non-Asymptotic Analysis For DDIM-Type Samplers

Sitan Chen, Giannis Daras, Alexandros G. Dimakis

Comments: 29 pages

Subjects: Machine Learning (cs.LG); Statistics Theory (math.ST); Machine Learning (stat.ML)

[187] arXiv:2303.03393 (cross-list from cs.LG) [pdf, other]

Title: Interpretable Architecture Neural Networks for Function Visualization

Shengtong Zhang, Daniel W. Apley

Subjects: Machine Learning (cs.LG); Human-Computer Interaction (cs.HC); Methodology (stat.ME); Machine Learning (stat.ML)

[188] arXiv:2303.03544 (cross-list from cs.LG) [pdf, other]

Title: Expressivity of Shallow and Deep Neural Networks for Polynomial Approximation

Itai Shapira

Comments: This paper contains 14 pages and no figures. This is a revised version with significant changes in sections 4 to clarify the proof of the main result

Subjects: Machine Learning (cs.LG); Machine Learning (stat.ML)

[189] arXiv:2303.03576 (cross-list from stat.CO) [pdf, other]

Title: A Survey of Numerical Algorithms that can Solve the Lasso Problems

Yujie Zhao, Xiaoming Huo

Subjects: Computation (stat.CO); Machine Learning (stat.ML)

[190] arXiv:2303.03589 (cross-list from math.ST) [pdf, other]

Title: Towards a Complete Analysis of Langevin Monte Carlo: Beyond Poincaré Inequality

Alireza Mousavi-Hosseini, Tyler Farghly, Ye He, Krishnakumar Balasubramanian, Murat A. Erdogdu

Comments: 28 pages. To appear in the Conference on Learning Theory (COLT) 2023

Subjects: Statistics Theory (math.ST); Probability (math.PR); Machine Learning (stat.ML)

[191] arXiv:2303.03891 (cross-list from math.OC) [pdf, other]

Title: Margin theory for the scenario-based approach to robust optimization in high dimension

Fabien Lauer (ABC)

Subjects: Optimization and Control (math.OC); Machine Learning (stat.ML)

[192] arXiv:2303.03900 (cross-list from math.OC) [pdf, html, other]

Title: Nash Equilibria, Regularization and Computation in Optimal Transport-Based Distributionally Robust Optimization

Soroosh Shafiee, Liviu Aolaritei, Florian Dörfler, Daniel Kuhn

Subjects: Optimization and Control (math.OC); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)

[193] arXiv:2303.03950 (cross-list from cs.LG) [pdf, other]

Title: On the existence of optimal shallow feedforward networks with ReLU activation

Steffen Dereich, Sebastian Kassing

Comments: arXiv admin note: substantial text overlap with arXiv:2302.14690

Subjects: Machine Learning (cs.LG); Numerical Analysis (math.NA); Optimization and Control (math.OC); Machine Learning (stat.ML)

[194] arXiv:2303.04012 (cross-list from cs.LG) [pdf, other]

Title: Exploration via Epistemic Value Estimation

Simon Schmitt, John Shawe-Taylor, Hado van Hasselt

Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)

[195] arXiv:2303.04040 (cross-list from cs.LG) [pdf, html, other]

Title: Uncertainty Quantification of Spatiotemporal Travel Demand with Probabilistic Graph Neural Networks

Qingyi Wang, Shenhao Wang, Dingyi Zhuang, Haris Koutsopoulos, Jinhua Zhao

Subjects: Machine Learning (cs.LG); Applications (stat.AP); Machine Learning (stat.ML)

[196] arXiv:2303.04124 (cross-list from physics.chem-ph) [pdf, other]

Title: Wigner kernels: body-ordered equivariant machine learning without a basis

Filippo Bigi, Sergey N. Pozdnyakov, Michele Ceriotti

Subjects: Chemical Physics (physics.chem-ph); Machine Learning (cs.LG); Machine Learning (stat.ML)

[197] arXiv:2303.04143 (cross-list from cs.LG) [pdf, other]

Title: Can We Scale Transformers to Predict Parameters of Diverse ImageNet Models?

Boris Knyazev, Doha Hwang, Simon Lacoste-Julien

Comments: ICML 2023, camera ready (7 tables with extra results added), code and models are at this https URL

Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (stat.ML)

[198] arXiv:2303.04145 (cross-list from cs.LG) [pdf, other]

Title: Benign Overfitting for Two-layer ReLU Convolutional Neural Networks

Yiwen Kou, Zixiang Chen, Yuanzhou Chen, Quanquan Gu

Comments: 45 pages, 3 figures, 2 tables. In ICML 2023

Subjects: Machine Learning (cs.LG); Optimization and Control (math.OC); Machine Learning (stat.ML)

[199] arXiv:2303.04156 (cross-list from cs.LG) [pdf, other]

Title: Computing with Categories in Machine Learning

Eli Sennesh, Tom Xu, Yoshihiro Maruyama

Comments: Submitted to AGI 2023

Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Category Theory (math.CT); Machine Learning (stat.ML)

[200] arXiv:2303.04209 (cross-list from cs.LG) [pdf, other]

Title: Causal Dependence Plots

Joshua R. Loftus, Lucius E. J. Bynum, Sakina Hansen

Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)