Worst-Case

Follow this topic Ask about this topic

Momentum

6 papers in the last four weeks, with none the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 36

All topics

Tight Transition Time Bounds for Separable Logistic Regression at the Edge of Stability
Oct 1, 2026Haodong Wen, Kaiyue Wen, Jiaye TengLogistic Regression Optimal Sample Complexity
Stanford University · Shanghai University of Finance and Economics
The Row Normalization Puzzle in Muon
Sep 30, 2026Jiayu Zhang, Tianyi LinMuon Normalization
Department of Industrial Engineering and Operations Research Columbia University
Kolmogorov-Arnold Classifier Systems as Universal Approximators
Sep 29, 2026Hiroki Shiraishi, Hisao Ishibuchi, Masaya NakataKolmogorov-Arnold Networks Classifier
Graduate School of Engineering Science, Yokohama National University, Yokohama 240-8501, Japan · Digital Healthcare Research Department, Hitachi, Ltd., Tokyo 185-8601, Japan · Department of Computer Science and Engineering, Southern University of Science and Technology, Shenzhen 518055, China +1
Worst-Case Hidden-Vehicle Trajectory Search in Spatiotemporal Occlusion Regions
Sep 17, 2026Ruichen Tan, Zengxiang Lei, Satish UkkusuriAutonomous Driving Robust Trajectory
Lyles School of Civil and Construction Engineering, Purdue University, West Lafayette, IN 47907, USA
Counterfactual Marginalisation: Framework for Evaluating Robustness to Nuisance Variables
Sep 12, 2026Yasin Ibrahim, Hermione Warr, Robin J. Evans +1Minority Worst-Case
Department of Engineering Science, University of Oxford, Oxford, UK · Department of Statistics, University of Oxford, Oxford, UK
A Sharp Barrier for Consistent Submodular Maximization: Any Improvement over $2-\sqrt{2}$ Entails Exponential Queries or Linear Recourse
Sep 9, 2026Shi Fu, Qixin Zhang, Dacheng TaoSubmodular Functions First Order Oracle Complexity
Nanyang Technological University, Singapore
Kernel Methods for Refined Prophet Inequalities
Aug 9, 2026Patrick Loiseau, Mathieu Molina, Vianney Perchet +2Concentration Bounds Worst-Case
FairPlay Joint Team, Inria · Tel-Aviv University · FairPlay Joint Team, CREST, ENSAE, IP Paris, CRITEO AI Team +2
Stability of Ranking-dependent Pair-wise Comparison Patterns in the Analytic Hierarchy Process
Aug 6, 2026Vitaliy Tsyganok, Sergii Kadenko, Oleh AndriichukComparison Worst-Case
Institute for Information Recording National Academy of Sciences of Ukraine Kyiv, Ukraine 03113
Quality Control Algorithms for Pattern Counting
Aug 4, 2026Cassandra Marcussen, Ronitt Rubinfeld, Madhu SudanO(\Bar{K}\Log N)$Worst-Case
School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts, USA. · Computer Science and Artificial Intelligence Laboratory, MIT, Cambridge, Massachusetts, USA.
Tight Worst-Case Bounds for the Smallest Eigenvalue of ReLU NTK Gram Matrices
Aug 4, 2026Zhao SongRectified Linear Unit Spectral Decomposition
Learning from Local Walks on Dynamic Graphs with Bandit Feedback
Jul 12, 2026Sourav Chakraborty, Amit Kiran Rege, Claire Monteleoni +1Stochastic Multi-Armed Bandits Bandits
Department of Computer Science, University of Colorado, Boulder, CO 80309, USA · INRIA, Paris, France
How AI settled the complexity of the oldest SGD algorithm
Jun 28, 2026Michał Dereziński, Xiaoyu DongStochastic Gradient Descent Worst-Case
University of Michigan · National University of Singapore
Graph Dimensionality Reduction for Contextual Bandits: Structure-Specific Regret Bounds under Approximate Smoothness and Noisy Eigenspaces
Jun 26, 2026Joyanta Jyoti Mondal, Ibne Farabi Shihab, Anuj SharmaContextual Bandit Framework Dimensionality Reduction
Department of Computer and Information Sciences, University of Delaware, USA · Department of Computer Science, Iowa State University, USA · Department of Civil, Construction & Environmental Engineering, Iowa State University, USA
Exponentially many initializations to avoid barren plateaus
Jun 16, 2026Ankit Kulshrestha, Ricard Puig, Diego García-Martín +4Worst-Case Concentration Bounds
Fujitsu Research of America, Santa Clara, CA 95054, USA · Institute of Physics, Ecole Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland · Centre for Quantum Science and Engineering, Ecole Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland +4
The Program Is Still There: A Conservation Law for Program Discovery
Jun 11, 2026Jorge Miguel SilvaSearch Algorithms Worst-Case
Institute of Electronics and Informatics Engineering of Aveiro (IEETA) and Department of Electronics, Telecommunications and Informatics (DETI), University of Aveiro, 3810-193 Aveiro, Portugal
Measuring Model Robustness via Fisher Information: Spectral Bounds, Theoretical Guarantees, and Practical Algorithms
Jun 3, 2026Chong Zhang, Xiang Li, Jia Wang +2Robustness Verification Fisher Information Matrix
Xi’an Jiaotong-Liverpool University · The Chinese University of Hong Kong
Policy Gradient for Continuous-Time Robust Markov Decision Processes
Jun 3, 2026Tanya Veeravalli, David M. Bossens, Atsushi NitandaMarkov Decision Processes Policy Gradient
Centre for Frontier AI Research, Agency for Science, Technology and Research (A*STAR) · Institute of High Performance Computing, Agency for Science, Technology and Research (A*STAR) · College of Computing and Data Science, Nanyang Technological University
A Practical Upper Bound on Selection Bias Effects in Medical Prediction Models
May 30, 2026Kara Liu, Maggie Wang, Russ B. AltmanSelection Bias Model Selection
Stanford University · Stanford, California, USA
On Language Generation in the Limit with Bounded Memory
May 28, 2026Jon Kleinberg, Anay Mehrotra, Amin Saberi +1Bounded-Memory Learnability
Cornell University · Stanford University · Google Research
Variance-Adaptive Optimal Algorithm for Reinforcement Learning with Multinomial Logit Function Approximation
May 27, 2026Wonyoung Kim, Min-Hwan Oh, Garud Iyengar +1Markov Decision Processes Optimal Policies
Chung-Ang University · Seoul National University · Columbia University
When Seekers Are Hard to Help: Evaluating Emotional Support Dialogue Systems in Worst-Case Interactions
May 27, 2026Jiajie Yang, Yangchun Li, Guanyi Chen +3Emotional Support Conversation Task User Simulation
Hubei Provincial Key Laboratory of Artificial Intelligence and Smart Learning · National Language Resources Monitoring and Research Center for Network Media · School of Computer Science, Central China Normal University +2
When Dynamics Shift, Robust Task Inference Wins: Offline Imitation Learning with Behavior Foundation Models Revisited
May 16, 2026Rishabh Agrawal, Rahul Jain, Ashutosh NayyarImitation Learning Model-Based Reinforcement Learning
University of Southern California
Orthologic for SAT Solving
May 14, 2026Vladislas de Haldat, Simon Guilloud, Viktor KunčakSatisfiability First Order Oracle Complexity
1EPFL
Worst-Case Regret Bounds for Combinatorial Thompson Sampling in Sleeping Semi-Bandits
May 10, 2026Zhiming Huang, Bingshan Hu, Jianping PanThompson Sampling Stochastic Multi-Armed Bandits
Paul G. Allen School of Computer Science & Engineering, University of Washington, WA, USA · Department of Computer Science, University of British Columbia, BC, Canada · Department of Computer Science, University of Victoria, BC, Canada
Instance-Adaptive Online Multicalibration
May 10, 2026Zhiming Huang, Jamie Morgenstern, Aaron Roth +1Learning-Augmented Algorithms Worst-Case
Paul G. Allen School of Computer Science and Engineering, University of Washington · Department of Computer and Information Sciences, University of Pennsylvania
Bayesian Sensitivity of Causal Inference Estimators under Evidence-Based Priors
May 8, 2026Nikita Dhawan, Daniel Shen, Leonardo Cotta +1Causal Inferences Bayesian
University of Toronto Vector Institute · University of Waterloo · Vector Institute
Worst-Case Discovery and Runtime Protection for RL-Based Network Controllers
May 6, 2026Hongyu Hè, Minhao Jin, Maria ApostolakiReinforcement Learning Control Worst-Case
Princeton
Lyapunov-Certified Direct Switching Theory for Q-Learning
Apr 21, 2026Donghwan LeeQ-Learning Lyapunov Function
Department of Electrical Engineering Korea Advanced Institute of Science and Technology (KAIST) Daejeon 34141, South Korea
The Cost of Relaxation: Evaluating the Error in Convex Neural Network Verification
Apr 20, 2026Merkouris Papamichail, Konstantinos Varsos, Giorgos Flouris +1Neural Network Verification Rectified Linear Unit Networks
Foundation for Research and Technology – Hellas, Heraklion, Greece · University of Crete, Heraklion, Greece · Catalan Institution for Research and Advanced Studies, Barcelona, Spain +1
Learning Unanimously Acceptable Lotteries via Queries
Apr 19, 2026Davin Choo, Paul W. Goldberg, Nicholas TehLotteries Majority Voting
Harvard University, USA · University of Oxford, UK
Algorithms for Deciding the Safety of States in Fully Observable Non-deterministic Problems: Technical Report
Mar 16, 2026Johannes Schmalz, Chaahat JainProbabilistic Safety Worst-Case
Saarland University, Germany
Poly-attention: a general scheme for higher-order self-attention
Feb 2, 2026Sayak Chakrabarti, Toniann Pitassi, Josh AlmanLinear Attention Transformer Architectures
Computer Science, Columbia University, New York, NY 10027.
GraphToxin: Reconstructing Full Unlearned Graphs from Graph Unlearning
Nov 14, 2025Ying Song, Balaji PalanisamyMembership Inference Attacks Unlearnable Examples
University of Pittsburgh Pittsburgh, PA, USA
Are Targeted Data Poisoning Attacks as Effective as We Think?
Sep 8, 2025William Xu, Chenyu Zhang, Yihan Wang +5Poisoning Hard
Waabi AI · University of Waterloo · Carnegie Mellon University +3
Random Walk Learning and the Pac-Man Attack
Aug 1, 2025Xingran Chen, Parimal Parag, Rohit Bhagat +2Decentralized Learning Bounded Adversary
Department of Electrical and Computer Engineering, Rutgers University, Piscataway Township, NJ 08854, USA · Department of Electrical Communication Engineering, Indian Institute of Science, Bangalore, Karnataka 560012, India
Estimating Uncertain Spatial Relationships in Robotics
Mar 27, 2013Randall Smith, Matthew Self, Peter CheesemanState Estimation Map Uncertainty
SRI International 333 Ravenswood Avenue Menlo Park, California 94025 · General Motors Research Laboratories Warren, Michigan · UC Berkeley +1

Oct 1, 2026, cs.LGJ/K move · Enter open · S save

Tight Transition Time Bounds for Separable Logistic Regression at the Edge of Stability

Haodong Wen, Kaiyue Wen, Jiaye Teng

Stanford University · Shanghai University of Finance and Economics

Logistic Regression Optimal Sample Complexity Gradient Descent Step Accuracy Worst-Case Edge

Open paper AskPDF arXiv

We study logistic regression on linearly separable data under gradient descent with a large constant stepsize

η

. Such dynamics may exhibit a characteristic Edge of Stability phenomenon, in which the loss initially oscillates before transitioning to a stable phase of monotone decrease. Existing work provides a tight

Θ(1)

bound in dimension

d=2

η\to \infty

and conjectures a bound independent of

η

in arbitrary dimensions

d\geq 2

. In this paper, we disprove this conjecture by showing that, for every fixed sample size

n\geq 2

and sufficiently small margin

γ

, the worst-case transition time is

Θ\!\left((\logη)^{\min\{n-2,d-2\}}\right)

uniformly over

d\geq2

. The key challenge in establishing a tight bound is that the sample contributing most strongly to the gradient can change repeatedly across iterations. To address this issue, we control such changes by induction on dimension and sample size, and construct matching hard instances.

Momentum

Related topics

Latest papers 36