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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multitask Reinforcement Learning is a promising way to obtain models withbetter performance, generalisation, data efficiency, and robustness. Mostexisting work is limited to compatible settings, where the state and actionspace dimensions are the same across tasks. Graph Neural Networks (GNN) are oneway to address incompatible environments, because they can process graphs ofarbitrary size. They also allow practitioners to inject biases encoded in thestructure of the input graph. Existing work in graph-based continuous controluses the physical morphology of the agent to construct the input graph, i.e.,encoding limb features as node labels and using edges to connect the nodes iftheir corresponded limbs are physically connected. In this work, we present aseries of ablations on existing methods that show that morphologicalinformation encoded in the graph does not improve their performance. Motivatedby the hypothesis that any benefits GNNs extract from the graph structure areoutweighed by difficulties they create for message passing, we also proposeAmorpheus, a transformer-based approach. Further results show that, whileAmorpheus ignores the morphological information that GNNs encode, itnonetheless substantially outperforms GNN-based methods that use themorphological information to define the message-passing scheme.

My Body is a Cage: the Role of Morphology in Graph-Based Incompatible Control