paper-with-me

Papers

Incorporating Human Flexibility through Reward Preferences in Human-AI Teaming

2023-12-21 · Siddhant Bhambri, Mudit Verma, Upasana Biswas, Anil Murthy, Subbarao Kambhampati

Preference-based Reinforcement Learning (PbRL) has made significant strides in single-agent settings, but has not been studied for multi-agent frameworks. On the other hand, modeling cooperation between multiple agents, specifically, Human-AI Teaming settings while ensuring successful task completion is a challenging problem. To this end, we perform the first investigation of multi-agent PbRL by extending single-agent PbRL to the two-agent teaming settings and formulate it as a Human-AI PbRL Cooperation Game, where the RL agent queries the human-in-the-loop to elicit task objective and human's preferences on the joint team behavior. Under this game formulation, we first introduce the notion of Human Flexibility to evaluate team performance based on if humans prefer to follow a fixed policy or adapt to the RL agent on the fly. Secondly, we study the RL agent's varying access to the human policy. We highlight a special case along these two dimensions, which we call Specified Orchestration, where the human is least flexible and agent has complete access to human policy. We motivate the need for taking Human Flexibility into account and the usefulness of Specified Orchestration through a gamified user study. We evaluate state-of-the-art PbRL algorithms for Human-AI cooperative setups through robot locomotion based domains that explicitly require forced cooperation. Our findings highlight the challenges associated with PbRL by varying Human Flexibility and agent's access to the human policy. Finally, we draw insights from our user study and empirical results, and conclude that Specified Orchestration can be seen as an upper bound PbRL performance for future research in Human-AI teaming scenarios.

📄 PDF Abstract BibTeX arXiv:2312.14292

Code (0)

등록된 구현이 없습니다.

Tasks

Benchmarkingreinforcement-learning

Similar Papers 제목 키워드 기반

Adaptive Preference Scaling for Reinforcement Learning with Human Feedback

2024-06-04 · Ilgee Hong, Zichong Li, Alexander Bukharin, Yixiao Li 외

Reinforcement learning from human feedback (RLHF) is a prevalent approach to align AI systems with human values by learning rewards from human preference data. Due to various reasons, however, such data typically takes t…

reinforcement-learningReinforcement LearningText Generation

From Demonstrations to Rewards: Alignment Without Explicit Human Preferences

2025-03-15 · Siliang Zeng, Yao Liu, Huzefa Rangwala, George Karypis 외

One of the challenges of aligning large models with human preferences lies in both the data requirements and the technical complexities of current approaches. Predominant methods, such as RLHF, involve multiple steps, ea…

reinforcement-learningReinforcement Learning

Online Policy Learning from Offline Preferences

2024-03-15 · Guoxi Zhang, Han Bao, Hisashi Kashima

In preference-based reinforcement learning (PbRL), a reward function is learned from a type of human feedback called preference. To expedite preference collection, recent works have leveraged \emph{offline preferences}, …

continuous-controlContinuous Control

VRM: Teaching Reward Models to Understand Authentic Human Preferences

2026-03-05 · Biao Liu, Ning Xu, Junming Yang, Hao Xu 외 arxiv

Large Language Models (LLMs) have achieved remarkable success across diverse natural language tasks, yet the reward models employed for aligning LLMs often encounter challenges of reward hacking, where the approaches pre…

Test-Time Preference Optimization: On-the-Fly Alignment via Iterative Textual Feedback

2025-01-22 · Yafu Li, Xuyang Hu, Xiaoye Qu, Linjie Li 외

Large language models (LLMs) demonstrate impressive performance but lack the flexibility to adapt to human preferences quickly without retraining. In this work, we introduce Test-time Preference Optimization (TPO), a fra…

Instruction Following