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

We address the problem of teaching a deep reinforcement learning (RL) agentto follow instructions in multi-task environments. Instructions are expressedin a well-known formal language -- linear temporal logic (LTL) -- and canspecify a diversity of complex, temporally extended behaviours, includingconditionals and alternative realizations. Our proposed learning approachexploits the compositional syntax and the semantics of LTL, enabling our RLagent to learn task-conditioned policies that generalize to new instructions,not observed during training. To reduce the overhead of learning LTL semantics,we introduce an environment-agnostic LTL pretraining scheme which improvessample-efficiency in downstream environments. Experiments on discrete andcontinuous domains target combinatorial task sets of up to $\sim10^{39}$ uniquetasks and demonstrate the strength of our approach in learning to solve(unseen) tasks, given LTL instructions.

LTL2Action: Generalizing LTL Instructions for Multi-Task RL