Convergence Behavior

Convergence behavior in machine learning and related fields investigates the conditions under which iterative algorithms reach a stable solution, a crucial aspect for algorithm design and reliability. Current research focuses on analyzing convergence in diverse settings, including non-convex optimization (e.g., using stochastic gradient descent with momentum and time-window analysis), multi-agent systems (e.g., in reinforcement learning and game theory), and weakly supervised learning. Understanding convergence properties is vital for improving algorithm efficiency, ensuring stability in complex systems, and developing robust solutions for applications ranging from robotics and search engine marketing optimization to large language model training.

Papers

November 7, 2024

Stem-OB: Generalizable Visual Imitation Learning with Stem-Like Convergent Observation through Diffusion Inversion
Kaizhe Hu, Zihang Rui, Yao He, Yuyao Liu, Pu Hua, Huazhe Xu
Image Diffusion Model Image Inversion STEM Education Convergence Behavior Diffusion Inversion

October 9, 2024

CoBa: Convergence Balancer for Multitask Finetuning of Large Language Models
Zi Gong, Hang Yu, Cong Liao, Bingchang Liu, Chaoyu Chen, Jianguo Li
Large Language Model Multi Task Learning Multi Task Convergence Behavior Convergence Performance Multitask Finetuning

August 12, 2024

Stable-BC: Controlling Covariate Shift with Stable Behavior Cloning
Shaunak A. Mehta, Yusuf Umut Ciftci, Balamurugan Ramachandran, Somil Bansal, Dylan P. Losey
Feature Imitation Behavior Cloning Covariate Shift Stable Code Convergence Behavior Learning Robot

May 27, 2024

Convergence of SGD with momentum in the nonconvex case: A time window-based analysis
Junwen Qiu, Bohao Ma, Andre Milzarek
Stochastic Gradient Descent Early Stage Convergence Convergence Analysis Residual Momentum General Nonconvex Local Convergence Convergence Behavior

May 25, 2024

Convergence Behavior of an Adversarial Weak Supervision Method
Steven An, Sanjoy Dasgupta
Weak Supervision Data Labeling Label Supervision Convergence Behavior

March 26, 2024

Compressed Federated Reinforcement Learning with a Generative Model
Ali Beikmohammadi, Sarit Khirirat, Sindri Magnússon
Reinforcement Learning Federated Learning Generative Modeling Q$ Learning Convergence Behavior Error Bounded Lossy

February 2, 2024

A Survey on Large Language Model Hallucination via a Creativity Perspective
Xuhui Jiang, Yuxing Tian, Fengrui Hua, Chengjin Xu, Yuanzhuo Wang, Jian Guo
Timely Survey Sustained Creativity Language Model Hallucination Creativity Assessment Convergence Behavior Creative Task

January 29, 2024

Strategic Usage in a Multi-Learner Setting
Eliot Shekhtman, Sarah Dean
General Strategy Online Learning Algorithm User Intent Strategic Behavior Convergence Behavior

August 18, 2023

Baird Counterexample is Solved: with an example of How to Debug a Two-time-scale Algorithm
Hengshuai Yao
Convergence Rate Single Example Time Scale Policy Algorithm Convergence Behavior Counter Example

August 9, 2023

Variations on the Reinforcement Learning performance of Blackjack
Avish Buramdoyal, Tim Gebbie
Category Wise Variation Convergence Behavior Poker Hand Reinforcement Learning Performance

July 26, 2023

Stability of Multi-Agent Learning: Convergence in Network Games with Many Players
Aamal Hussain, Dan Leonte, Francesco Belardinelli, Georgios Piliouras
Early Stage Convergence Core Stability Multi Agent Learning Zero Sum Game Two Player Stable Learning Convergence Behavior

July 20, 2023

On the Convergence of Bounded Agents
David Abel, André Barreto, Hado van Hasselt, Benjamin Van Roy, Doina Precup, Satinder Singh
Early Stage Convergence Reinforcement Learning Problem Learning Problem Convergence Behavior Bounded Rational Future Behavior

June 30, 2023

TD Convergence: An Optimization Perspective
Kavosh Asadi, Shoham Sabach, Yao Liu, Omer Gottesman, Rasool Fakoor
Reinforcement Learning Iterative Optimization Convergence Behavior Quadratic Loss

June 21, 2023

AdCraft: An Advanced Reinforcement Learning Benchmark Environment for Search Engine Marketing Optimization
Maziar Gomrokchi, Owen Levin, Jeffrey Roach, Jonah White
Reinforcement Learning Deep Reinforcement Learning Reinforcement Learning Algorithm Reinforcement Learning Benchmark Convergence Behavior Sponsored Search

January 7, 2023

Investigations on convergence behaviour of Physics Informed Neural Networks across spectral ranges and derivative orders
Mayank Deshpande, Siddharth Agarwal, Vukka Snigdha, Arya Kumar Bhattacharya
Loss Function Physic Informed Neural Network Neural Tangent Kernel Comprehensive Investigation Spectral Bias Convergence Behavior Derivative Process Spectral Band

December 22, 2022

Stochastic analysis of the Elo rating algorithm in round-robin tournaments
Daniel Gomes de Pinho Zanco, Leszek Szczecinski, Eduardo Vinicius Kuhn, Rui Seara
Practical Algorithm Stochastic Way Convergence Behavior Elo Rating Round Robin Tournament

October 29, 2022

Recursive Reasoning in Minimax Games: A Level $k$ Gradient Play Method
Zichu Liu, Lacra Pavel
Generative Adversarial Network Practical Method Gradient Norm GAN Training Different Level Convergence Behavior Recursive Thinking Minimax Game

October 28, 2022

Motion Planning using Reactive Circular Fields: A 2D Analysis of Collision Avoidance and Goal Convergence
Marvin Becker, Johannes Köhler, Sami Haddadin, Matthias A. Müller
Motion Planning Collision Avoidance Trajectory Planning Trajectory Generation Motion Planning Algorithm Global Path Planning Convergence Behavior 3D Field

August 30, 2022

Super-model ecosystem: A domain-adaptation perspective
Fengxiang He, Dacheng Tao
Domain Adaptation Large Scale Model Convergence Behavior Model Paradigm

April 14, 2022

Leveraging convergence behavior to balance conflicting tasks in multi-task learning
Angelica Tiemi Mizuno Nakamura, Denis Fernando Wolf, Valdir Grassi
Strong Generalization Multi Task Learning Generalization Performance New Task Generalization Capability Convergence Behavior Simultaneous Learning