paper-with-me

홈 › Papers

Is a Good Foundation Necessary for Efficient Reinforcement Learning? The Computational Role of the Base Model in Exploration

2025-03-10 · Dylan J. Foster, Zakaria Mhammedi, Dhruv Rohatgi

Language model alignment (or, reinforcement learning) techniques that leverage active exploration -- deliberately encouraging the model to produce diverse, informative responses -- offer the promise of super-human capabilities. However, current understanding of algorithm design primitives for computationally efficient exploration with language models is limited. To better understand how to leverage access to powerful pre-trained generative models to improve the efficiency of exploration, we introduce a new computational framework for RL with language models, in which the learner interacts with the model through a sampling oracle. Focusing on the linear softmax model parameterization, we provide new results that reveal the computational-statistical tradeoffs of efficient exploration: 1. Necessity of coverage: Coverage refers to the extent to which the pre-trained model covers near-optimal responses -- a form of hidden knowledge. We show that coverage, while not necessary for data efficiency, lower bounds the runtime of any algorithm in our framework. 2. Inference-time exploration: We introduce a new algorithm, SpannerSampling, which obtains optimal data efficiency and is computationally efficient whenever the pre-trained model enjoys sufficient coverage, matching our lower bound. SpannerSampling leverages inference-time computation with the pre-trained model to reduce the effective search space for exploration. 3. Insufficiency of training-time interventions: We contrast the result above by showing that training-time interventions that produce proper policies cannot achieve similar guarantees in polynomial time. 4. Computational benefits of multi-turn exploration: Finally, we show that under additional representational assumptions, one can achieve improved runtime (replacing sequence-level coverage with token-level coverage) through multi-turn exploration.

📄 PDF Abstract BibTeX arXiv:2503.07453

Code (0)

등록된 구현이 없습니다.

Tasks

Efficient Exploration

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

Reinforcement Learning for Accelerated Aerodynamic Shape Optimisation

2025-07-23 · Florian Sobieczky, Alfredo Lopez, Erika Dudkin, Christopher Lackner 외 arxiv

We introduce a reinforcement learning (RL) based adaptive optimization algorithm for aerodynamic shape optimization focused on dimensionality reduction. The form in which RL is applied here is that of a surrogate-based, …

Dimensionality ReductionReinforcement LearningFeature Importance

Learning Norms via Natural Language Teachings

2022-01-20 · Taylor Olson, Ken Forbus

To interact with humans, artificial intelligence (AI) systems must understand our social world. Within this world norms play an important role in motivating and guiding agents. However, very few computational theories fo…

You Don't Need All Attentions: Distributed Dynamic Fine-Tuning for Foundation Models

2025-04-16 · Shiwei Ding, Lan Zhang, Zhenlin Wang, Giuseppe Ateniese 외

Fine-tuning plays a crucial role in adapting models to downstream tasks with minimal training efforts. However, the rapidly increasing size of foundation models poses a daunting challenge for accommodating foundation mod…

AllDistributed Computingparameter-efficient fine-tuning

A Memory Efficient Deep Reinforcement Learning Approach For Snake Game Autonomous Agents

2023-01-27 · Md. Rafat Rahman Tushar, Shahnewaz Siddique

To perform well, Deep Reinforcement Learning (DRL) methods require significant memory resources and computational time. Also, sometimes these systems need additional environment information to achieve a good reward. Howe…

Deep Reinforcement Learningreinforcement-learningReinforcement Learning (RL)

Exploration is Harder than Prediction: Cryptographically Separating Reinforcement Learning from Supervised Learning

2024-04-04 · Noah Golowich, Ankur Moitra, Dhruv Rohatgi

Supervised learning is often computationally easy in practice. But to what extent does this mean that other modes of learning, such as reinforcement learning (RL), ought to be computationally easy by extension? In this w…

regressionReinforcement Learning (RL)