1

Partially Observable Task and Motion Planning with Uncertainty and Risk Awareness

Aidan Curtis, George Matheos, Nishad Gothoskar, Vikash K. Mansinghka, Joshua Tenenmaum, Tomás Lozano-Pérez, Leslie Pack Kaelbling

SMCP3: SMC with Probabilistic Program Proposals

This paper introduces SMCP3, a method for automatically implementing custom sequential Monte Carlo samplers for inference in …

Alexander K. Lew, George Matheos, Tan Zhi-Xuan, Matin Ghavamizadeh, Nishad Gothoskar, Stuart Russell, Vikash K. Mansinghka

Brain computation as fast spiking neural Monte Carlo inference in probabilistic programs

George Matheos, Andrew D. Bolton, McCoy Becker, Cameron Freer, Vikash K. Mansinghka

Transforming Worlds: Automated Involutive MCMC for Open Universe Probabilistic Programs

Inference in open-universe probabilistic models can be challenging. We show how to automate a broad class of MCMC kernels for them, facilitating the development of domain-specific algorithms for inference about unknown objects & their relationships.

George Matheos, Alexander K. Lew, Matin Ghavamizadeh, Stuart Russell, Marco Cusumano-Towner, Vikash K. Mansinghka