Call for Papers: CoLLAs 2025 – The 4th Conference on Lifelong Learning Agents
CoLLAs 2025, the 4th Conference on Lifelong Learning Agents, is now accepting submissions for its annual gathering of researchers working on machine learning systems that can continually learn throughout their lifetime. The conference will be held at the University of Pennsylvania in Philadelphia, USA, and will focus on new theories, methodologies, applications, or insights into existing algorithms and benchmarks for learning in non-i.i.d. and non-stationary settings.
The 4th Conference on Lifelong Learning Agents (CoLLAs 2025) is now accepting submissions for its annual gathering of researchers working on machine learning systems that can continually learn throughout their lifetime. This year’s conference will be held at the University of Pennsylvania in Philadelphia, USA. We invite submissions that describe new theories, methodologies, applications, or insights into existing algorithms and benchmarks for learning in non-i.i.d. and non-stationary settings. Topics of submission include theory for continual/lifelong learning, continual learning paradigms, challenges in non-stationary learning, continual reinforcement learning, continual learning in large language models, knowledge transfer, non-stationary optimization, streaming learning, open-world learning, neuroscience-inspired continual/lifelong learning, and applications in control, robotics, and healthcare.
Submissions will be evaluated based on their novelty, technical quality, and potential impact. Experimental methods and results are expected to be reproducible, and authors are encouraged to make code and data available. We also encourage submissions of proof-of-concept work that puts forward novel ideas and demonstrates potential, as well as in-depth analysis of existing methods and concepts.
Tags: CoLLAs 2025, Conference on Lifelong Learning Agents, machine learning, continual learning, non-i.i.d., non-stationary settings, research, University of Pennsylvania, submissions, theory, methodologies, applications, algorithms, benchmarks, learning, neuroscience-inspired, control, robotics, healthcare