9 papers
Beyond Linear Attention: Softmax Transformers Implement In-Context Reinforcement Learning
Zixuan Xie, Xinyu Liu, Claire Chen +3
In-context reinforcement learning (ICRL) studies agents that, after pretraining, adapt to new tasks by conditioning on additional context without parameter updates. Existing theore…
MathlibPR: Pull Request Merge-Readiness Benchmark for Formal Mathematical Libraries
Zixuan Xie, Xinyu Liu, Shangtong Zhang
The ecosystem of Lean and Mathlib has become the de facto standard for large language model (LLM) assisted formal reasoning with remarkable successes in recent years. Those success…
Convergence and Emergence of In-Context Reinforcement Learning with Chain of Thought
Zixuan Xie, Xinyu Liu, Rohan Chandra +1
In-context reinforcement learning (ICRL) refers to the ability of RL agents to adapt to new tasks at inference time without parameter updates by conditioning on additional context.…
Almost Sure Convergence Rates of Stochastic Approximation and Reinforcement Learning via a Poisson-Moreau Drift
Xinyu Liu, Zixuan Xie, Shangtong Zhang
Establishing almost sure convergence rates for stochastic approximation and reinforcement learning under Markovian noise is a fundamental theoretical challenge. We make progress to…
On the Divergence of Differential Temporal Difference Learning without Local Clocks
David Antrobius, Shangtong Zhang
Learning rate is a critical component of reinforcement learning (RL). This work uses global and local clocks to distinguish two types of learning rates. The former is of the standa…
MathlibLemma: Folklore Lemma Generation and Benchmark for Formal Mathematics
Xinyu Liu, Zixuan Xie, Amir Moeini +5
While the ecosystem of Lean and Mathlib has enjoyed celebrated success in formal mathematical reasoning with the help of large language models (LLMs), the absence of many folklore…
Extensions of Robbins-Siegmund Theorem with Applications in Reinforcement Learning
Xinyu Liu, Zixuan Xie, Shangtong Zhang
The Robbins-Siegmund theorem establishes the convergence of stochastic processes that are almost supermartingales and is one of the most commonly used approaches for analyzing stoc…
Finite Sample Analysis of Linear Temporal Difference Learning with Arbitrary Features
Zixuan Xie, Xinyu Liu, Rohan Chandra +1
Linear TD() is one of the most fundamental reinforcement learning algorithms for policy evaluation. Previously, convergence rates are typically established under the assumption…
Linear -Learning Does Not Diverge in : Convergence Rates to a Bounded Set
Xinyu Liu, Zixuan Xie, Shangtong Zhang
-learning is one of the most fundamental reinforcement learning algorithms. It is widely believed that -learning with linear function approximation (i.e., linear -learning…