Calculus on MDPs: Potential Shaping as a Gradient

v1v2 (latest)

Calculus on MDPs: Potential Shaping as a Gradient

20 August 2022

Adam Gleave

ArXiv (abs)PDF HTML

Papers citing "Calculus on MDPs: Potential Shaping as a Gradient"

11 / 11 papers shown

Title
Preprocessing Reward Functions for Interpretability Erik Jenner Adam Gleave 121 8 0 25 Mar 2022
Invariance in Policy Optimisation and Partial Identifiability in Reward Learning Joar Skalse Matthew Farrugia-Roberts Stuart J. Russell Alessandro Abate Adam Gleave 56 47 0 14 Mar 2022
Dynamics-Aware Comparison of Learned Reward Functions Blake Wulfe Ashwin Balakrishna Logan Ellis Jean Mercat R. McAllister Adrien Gaidon 26 15 0 25 Jan 2022
Plan-Based Relaxed Reward Shaping for Goal-Directed Tasks Ingmar Schubert Ozgur S. Oguz Marc Toussaint OffRL 55 6 0 14 Jul 2021
Identifiability in inverse reinforcement learning Haoyang Cao Samuel N. Cohen Lukasz Szpruch 45 47 0 07 Jun 2021
Understanding Learned Reward Functions Eric J. Michaud Adam Gleave Stuart J. Russell XAI OffRL 67 34 0 10 Dec 2020
Reward Machines: Exploiting Reward Function Structure in Reinforcement Learning Rodrigo Toro Icarte Toryn Q. Klassen Richard Valenzano Sheila A. McIlraith OffRL 114 222 0 06 Oct 2020
Quantifying Differences in Reward Functions Adam Gleave Michael Dennis Shane Legg Stuart J. Russell Jan Leike OffRL 124 68 0 24 Jun 2020
Fine-Tuning Language Models from Human Preferences Daniel M. Ziegler Nisan Stiennon Jeff Wu Tom B. Brown Alec Radford Dario Amodei Paul Christiano G. Irving ALM 474 1,766 0 18 Sep 2019
Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations Daniel S. Brown Wonjoon Goo P. Nagarajan S. Niekum 76 358 0 12 Apr 2019
Deep reinforcement learning from human preferences Paul Christiano Jan Leike Tom B. Brown Miljan Martic Shane Legg Dario Amodei 216 3,364 0 12 Jun 2017

We use cookies and other tracking technologies to improve your browsing experience on our website, to show you personalized content and targeted ads, to analyze our website traffic, and to understand where our visitors are coming from. See our policy.