Excited to give my talk "New Design Decisions for Modern AI Agents" at #AAAI23 tomorrow (Sunday) morning at 8:30am in Ballroom ABC! An earlier (now somewhat out of date) version of the talk is here:
The first author Sebastiaan De Peuter does not follow twitter but is certainly wotrh talking with - I am proud of this paper, on Collaborative AI for design problems and sequential decision making more generally. @FCAI_fi #TuringAIFellows @idsai_uom
---
RT @FCAI_fi
Sunday Feb. 12 at #AAAI23 in Washington: AI assistance + automation for solving sequential decision problems.
Paper: Zero-Shot Assistance in Sequential Decision Problems (@sami…
https://twitter.com/FCAI_fi/status/1623780021847355392
“Sunday Feb. 12 at #AAAI23 in Washington: AI assistance + automation for solving sequential decision problems. Paper: Zero-Shot Assistance in Sequential Decision Problems (@samikaski et al.) https://t.co/9W23fh4TgU”
Are you at #AAAI23 this week? Check out these papers from FCAI:
*️⃣ Co-Imitation: Learning Design and Behaviour by Imitation (Rajani et al.) https://arxiv.org/abs/2209.01207 (project website https://sites.google.com/view/co-imitation)
*️⃣ Zero-Shot Assistance in Sequential Decision Problems (@samikaski et al.)
https://arxiv.org/abs/2202.07364
The co-adaptation of robots has been a long-standing research endeavour with the goal of adapting both body and behaviour of a system for a given task, inspired by the natural evolution of animals. Co-adaptation has the potential to eliminate costly manual hardware engineering as well as improve the performance of systems. The standard approach to co-adaptation is to use a reward function for optimizing behaviour and morphology. However, defining and constructing such reward functions is notoriously difficult and often a significant engineering effort. This paper introduces a new viewpoint on the co-adaptation problem, which we call co-imitation: finding a morphology and a policy that allow an imitator to closely match the behaviour of a demonstrator. To this end we propose a co-imitation methodology for adapting behaviour and morphology by matching state distributions of the demonstrator. Specifically, we focus on the challenging scenario with mismatched state- and action-spaces between both agents. We find that co-imitation increases behaviour similarity across a variety of tasks and settings, and demonstrate co-imitation by transferring human walking, jogging and kicking skills onto a simulated humanoid.