3 papers
cs.CL2023
A mixed policy to improve performance of language models on math problems
Gang Chen
When to solve math problems, most language models take a sampling strategy to predict next word according conditional probabilities. In the math reasoning step, it may generate wro…
cs.LG2023
Deep Metric Tensor Regularized Policy Gradient
Gang Chen, Victoria Huang
Policy gradient algorithms are an important family of deep reinforcement learning techniques. Many past research endeavors focused on using the first-order policy gradient informat…
cs.LG2022
MSRL: Distributed Reinforcement Learning with Dataflow Fragments
Huanzhou Zhu, Bo Zhao, Gang Chen +6
Reinforcement learning (RL) trains many agents, which is resource-intensive and must scale to large GPU clusters. Different RL training algorithms offer different opportunities for…