English
Related papers

Related papers: Reconciling Rewards with Predictive State Represen…

200 papers

Partially observable Markov decision processes (POMDPs) provide an elegant mathematical framework for modeling complex decision and planning problems in stochastic domains in which states of the system are observable only indirectly, via a…

Artificial Intelligence · Computer Science 2011-06-02 M. Hauskrecht

We present a method for a certain class of Markov Decision Processes (MDPs) that can relate the optimal policy back to one or more reward sources in the environment. For a given initial state, without fully computing the value function,…

Machine Learning · Computer Science 2018-06-12 Josh Bertram , Peng Wei

Reward models are crucial for aligning large language models (LLMs) with human values and intentions. Existing approaches follow either Generative (GRMs) or Discriminative (DRMs) paradigms, yet both suffer from limitations: GRMs typically…

Computation and Language · Computer Science 2026-03-03 Longze Chen , Lu Wang , Renke Shan , Ze Gong , Run Luo , Jiaming Li , Jing Luo , Qiyao Wang , Min Yang

Large language models (LLMs) inevitably make mistakes when performing step-by-step mathematical reasoning. Process Reward Models (PRMs) have emerged as a promising solution by evaluating each reasoning step. However, existing PRMs typically…

Computation and Language · Computer Science 2025-03-28 Shuaijie She , Junxiao Liu , Yifeng Liu , Jiajun Chen , Xin Huang , Shujian Huang

Inference-time scaling methods rely on Process Reward Models (PRMs), which are often poorly calibrated and overestimate success probabilities. We propose, to our knowledge, the first use of conditional optimal transport for calibrating…

Machine Learning · Computer Science 2026-05-13 Rachel Ma , Dylan Hadfield-Menell , Kristjan Greenewald

Multi-agent planning and reinforcement learning can be challenging when agents cannot see the state of the world or communicate with each other due to communication costs, latency, or noise. Partially Observable Stochastic Games (POSGs)…

Multiagent Systems · Computer Science 2024-12-20 Rafael F. Cunha , Jacopo Castellini , Johan Peralez , Jilles S. Dibangoye

Partially Observable Markov Decision Processes (POMDPs) are a natural and general model in reinforcement learning that take into account the agent's uncertainty about its current state. In the literature on POMDPs, it is customary to assume…

Machine Learning · Computer Science 2022-03-24 Noah Golowich , Ankur Moitra , Dhruv Rohatgi

Real-world sequential decision making problems commonly involve partial observability, which requires the agent to maintain a memory of history in order to infer the latent states, plan and make good decisions. Coping with partial…

Machine Learning · Computer Science 2022-02-09 Yonathan Efroni , Chi Jin , Akshay Krishnamurthy , Sobhan Miryoosefi

It is well known that for any finite state Markov decision process (MDP) there is a memoryless deterministic policy that maximizes the expected reward. For partially observable Markov decision processes (POMDPs), optimal memoryless policies…

Optimization and Control · Mathematics 2016-02-16 Guido Montufar , Keyan Ghazi-Zahedi , Nihat Ay

Process Reward Models (PRMs) emerge as a promising approach for process supervision in mathematical reasoning of Large Language Models (LLMs), which aim to identify and mitigate intermediate errors in the reasoning processes. However, the…

Computation and Language · Computer Science 2025-06-06 Zhenru Zhang , Chujie Zheng , Yangzhen Wu , Beichen Zhang , Runji Lin , Bowen Yu , Dayiheng Liu , Jingren Zhou , Junyang Lin

We study an approximation method for partially observed Markov decision processes (POMDPs) with continuous spaces. Belief MDP reduction, which has been the standard approach to study POMDPs requires rigorous approximation methods for…

Optimization and Control · Mathematics 2025-01-20 Ali Devran Kara , Erhan Bayraktar , Serdar Yuksel

Inverse reinforcement learning is the problem of inferring a reward function from an optimal policy or demonstrations by an expert. In this work, it is assumed that the reward is expressed as a reward machine whose transitions depend on…

Machine Learning · Computer Science 2025-10-23 Mohamad Louai Shehab , Antoine Aspeel , Necmiye Ozay

Partially observable Markov Decision Processes (POMDPs) are a standard model for agents making decisions in uncertain environments. Most work on POMDPs focuses on synthesizing strategies based on the available capabilities. However, system…

Artificial Intelligence · Computer Science 2024-07-12 Alyzia-Maria Konsta , Alberto Lluch Lafuente , Christoph Matheja

Partially observable Markov decision processes (POMDPs) are a fundamental model for sequential decision-making under uncertainty. However, many verification and synthesis problems for POMDPs are undecidable or intractable. Most prominently,…

Artificial Intelligence · Computer Science 2026-04-23 Nathanaël Fijalkow , Arka Ghosh , Roman Kniazev , Guillermo A. Pérez , Pierre Vandenhove

Given a Markov decision process (MDP), we seek to learn representations for a range of policies to facilitate behavior steering at test time. As policies of an MDP are uniquely determined by their occupancy measures, we propose modeling…

Machine Learning · Computer Science 2026-02-02 Beiming Li , Sergio Rozada , Alejandro Ribeiro

In inverse reinforcement learning (IRL), a learning agent infers a reward function encoding the underlying task using demonstrations from experts. However, many existing IRL techniques make the often unrealistic assumption that the agent…

Machine Learning · Computer Science 2023-01-04 Franck Djeumou , Christian Ellis , Murat Cubuktepe , Craig Lennon , Ufuk Topcu

Appropriate credit assignment for delay rewards is a fundamental challenge for reinforcement learning. To tackle this problem, we introduce a delay reward calibration paradigm inspired from a classification perspective. We hypothesize that…

Machine Learning · Computer Science 2021-08-26 Yixuan Liu , Hu Wang , Xiaowei Wang , Xiaoyue Sun , Liuyue Jiang , Minhui Xue

Partially Observable Markov Decision Processes (POMDPs) are a fundamental framework for decision-making under uncertainty and partial observability. Since in general optimal policies may require infinite memory, they are hard to implement…

Artificial Intelligence · Computer Science 2026-04-30 Muqsit Azeem , Debraj Chakraborty , Sudeep Kanav , Jan Kretinsky

A common setting in multitask reinforcement learning (RL) demands that an agent rapidly adapt to various stationary reward functions randomly sampled from a fixed distribution. In such situations, the successor representation (SR) is a…

Machine Learning · Computer Science 2023-09-08 Ted Moskovitz , Samo Hromadka , Ahmed Touati , Diana Borsa , Maneesh Sahani

Not having access to compact and meaningful representations is known to significantly increase the complexity of reinforcement learning (RL). For this reason, it can be useful to perform state representation learning (SRL) before tackling…

Machine Learning · Computer Science 2022-02-16 Astrid Merckling , Nicolas Perrin-Gilbert , Alex Coninx , Stéphane Doncieux