DeployedisContinual

Position: Deployed Reinforcement Learning should be Continual

ICML 2026 Paper Poster Presentation

Many deployments are continual RL problems approximated as non-continual. We term these ‘measurable deployments’.

May 2026 · Parnian Behdin*, Kevin Roice*, Golnaz Mesbahi
Abstract Models for Planning

Goal-Space Planning with Subgoal Models

JMLR 2024 Paper Poster Presentation

Background planning with option models allows for faster credit assignment and adaptation.

October 2024 · Chunlok Lo*, Kevin Roice*, Parham Mohammad Panahi*, Scott M. Jordan, Adam White, Gabor Michuz, Farzane Aminmansour, Martha White
RHapTor

RHapTor: Rendering Haptic Torques for Virtual Reality

SIGGRAPH 2022 Paper Code

Adding motor-driven discs to a virtual reality headset can induce noticeable gyroscopic torques on the wearer.

August 2022 · Kevin Roice, George A. Koulieris