AIConfPaper
← Search

Christian A. Schroeder De Witt

6 accepted papers

2022

Amortized Rejection Sampling in Universal Probabilistic Programming

Saeid Naderiparizi, Adam Scibior, Andreas Munk, Mehrdad Ghadiri, Atilim Gunes Baydin, Bradley J. Gram-Hansen +6
AISTATS 2022poster

Naive approaches to amortized inference in probabilistic programs with unbounded loops can produce estimators with infinite variance. This is particularly true of importance sampling inference in programs that explicitly include rejection sampling as part of the user-programmed generative procedure.…

Cited by 7Source ↗PDF ↗code ↗Scholar ↗
2022

Communicating via Markov Decision Processes

Samuel Sokota, Christian A Schroeder De Witt, Maximilian Igl, Luisa M Zintgraf, Philip Torr, Martin Strohmeier +3
ICML 2022spotlight

We consider the problem of communicating exogenous information by means of Markov decision process trajectories. This setting, which we call a Markov coding game (MCG), generalizes both source coding and a large class of referential games. MCGs also isolate a problem that is important in decentraliz…

Cited by 9Source ↗PDF ↗code ↗Scholar ↗
2022

Generalized Beliefs for Cooperative AI

Darius Muglich, Luisa M Zintgraf, Christian A Schroeder De Witt, Shimon Whiteson, Jakob Foerster
ICML 2022spotlight

Self-play is a common method for constructing solutions in Markov games that can yield optimal policies in collaborative settings. However, these policies often adopt highly-specialized conventions that make playing with a novel partner difficult. To address this, recent approaches rely on encoding…

Cited by 8Source ↗PDF ↗code ↗Scholar ↗
2022

Mirror Learning: A Unifying Framework of Policy Optimisation

Jakub Grudzien, Christian A Schroeder De Witt, Jakob Foerster
ICML 2022spotlight

Modern deep reinforcement learning (RL) algorithms are motivated by either the general policy improvement (GPI) or trust-region learning (TRL) frameworks. However, algorithms that strictly respect these theoretical frameworks have proven unscalable. Surprisingly, the only known scalable algorithms v…

Cited by 43Source ↗PDF ↗code ↗Scholar ↗
2022

Model-Free Opponent Shaping

Christopher Lu, Timon Willi, Christian A Schroeder De Witt, Jakob Foerster
ICML 2022spotlight

In general-sum games the interaction of self-interested learning agents commonly leads to collectively worst-case outcomes, such as defect-defect in the iterated prisoner’s dilemma (IPD). To overcome this, some methods, such as Learning with Opponent-Learning Awareness (LOLA), directly shape the lea…

Cited by 66Source ↗PDF ↗code ↗Scholar ↗
2021

Randomized Entity-wise Factorization for Multi-Agent Reinforcement Learning

Shariq Iqbal, Christian A Schroeder De Witt, Bei Peng, Wendelin Boehmer, Shimon Whiteson, Fei Sha
ICML 2021oral

Multi-agent settings in the real world often involve tasks with varying types and quantities of agents and non-agent entities; however, common patterns of behavior often emerge among these agents/entities. Our method aims to leverage these commonalities by asking the question: “What is the expected…

Cited by 83Source ↗PDF ↗code ↗Scholar ↗