1.2k citations · 3.6k across the 11 of their papers we have counts for
14 papers
Projected Distillation: On Learning to Solve Hard Problems from Close, Correct Targets
Shripad V. Deshmukh, Christian Bock, Kangrui Ruan +4
On-policy learning is central to post-training: training on a model's own generations limits distribution shift and helps preserve existing capabilities. On hard problems, however,…
ANTE: Anticipate, Then Execute -- Learning Internal World Models for Coding Agents from Verified Beliefs
Yi Ge, Kangrui Ruan, Jifan Chen +6
Rapid advances in large language models have made software engineering agents capable of tackling complex coding tasks. Within a containerized environment, a coding agent receives…
Reward-Aligned Distillation
Yilan Chen, Zian Su, Manish Prajapat +4
Reinforcement learning with verifiable rewards improves the reasoning capabilities of large language models, but remains computationally expensive and sample inefficient. On-policy…
Group Adaptive Clipping Policy Optimization
Sheng Jia, Xiao Wang, Shiva Prasad Kasiviswanathan +1
Group relative policy optimization for reinforcement learning with verifiable rewards (RLVR) typically uses a fixed importance-sampling (IS) ratio clipping boundary across all roll…
GenRec: An LLM-Backed Recommendation Ranker at Netflix
Ying Li, Shradha Sehgal, Arjun Rao +3
Large language models (LLMs) are reshaping recommender systems by enabling richer modeling of users, content, and context directly in natural language. At Netflix, we are exploring…
From Logs to Language: Learning Optimal Verbalization for LLM-Based Recommendation at Industry Scale
Yucheng Shi, Ying Li, Yu Wang +8
Large language models (LLMs) are promising backbones for generative recommender systems, yet a key challenge remains underexplored: verbalization, i.e., converting structured user…
Some Considerations on Learning to Explore via Meta-Reinforcement Learning
Bradly C. Stadie, Ge Yang, Rein Houthooft +5
We consider the problem of exploration in meta reinforcement learning. Two new meta reinforcement learning algorithms are suggested: E-MAML and E-. Results are present…
Evolved Policy Gradients
Rein Houthooft, Richard Y. Chen, Phillip Isola +4
We propose a metalearning approach for learning gradient-based reinforcement learning (RL) algorithms. The idea is to evolve a differentiable loss function, such that an agent, whi…
Parameter Space Noise for Exploration
Matthias Plappert, Rein Houthooft, Prafulla Dhariwal +6
Deep reinforcement learning (RL) methods generally engage in exploratory behavior through noise injection in the action space. An alternative is to add noise directly to the agent'…
#Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
Haoran Tang, Rein Houthooft, Davis Foote +6
Count-based exploration algorithms are known to perform near-optimally when used in conjunction with tabular reinforcement learning (RL) methods for solving small discrete Markov d…
InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
Xi Chen, Yan Duan, Rein Houthooft +3
This paper describes InfoGAN, an information-theoretic extension to the Generative Adversarial Network that is able to learn disentangled representations in a completely unsupervis…
Benchmarking Deep Reinforcement Learning for Continuous Control
Yan Duan, Xi Chen, Rein Houthooft +2
Recently, researchers have made significant progress combining the advances in deep learning for learning feature representations with reinforcement learning. Some notable examples…
VIME: Variational Information Maximizing Exploration
Rein Houthooft, Xi Chen, Yan Duan +3
Scalable and effective exploration remains a key challenge in reinforcement learning (RL). While there are methods with optimality guarantees in the setting of discrete state and a…
Integrated Inference and Learning of Neural Factors in Structural Support Vector Machines
Rein Houthooft, Filip De Turck
Tackling pattern recognition problems in areas such as computer vision, bioinformatics, speech or text recognition is often done best by taking into account task-specific statistic…