English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Natural behavior consists of dynamics that are complex and unpredictable,especially when trying to predict many steps into the future. While somesuccess has been found in building representations of behavior underconstrained or simplified task-based conditions, many of these models cannot beapplied to free and naturalistic settings where behavior becomes increasinglyhard to model. In this work, we develop a multi-task representation learningmodel for behavior that combines two novel components: (i) An action predictionobjective that aims to predict the distribution of actions over futuretimesteps, and (ii) A multi-scale architecture that builds separate latentspaces to accommodate short- and long-term dynamics. After demonstrating theability of the method to build representations of both local and globaldynamics in realistic robots in varying environments and terrains, we apply ourmethod to the MABe 2022 Multi-agent behavior challenge, where our model ranks1st overall and on all global tasks, and 1st or 2nd on 7 out of 9 frame-leveltasks. In all of these cases, we show that our model can build representationsthat capture the many different factors that drive behavior and solve a widerange of downstream tasks.

Relax, it doesn't matter how you get there: A new self-supervised approach for multi-timescale behavior analysis