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

Meta-training agents with memory has been shown to culminate in Bayes-optimalagents, which casts Bayes-optimality as the implicit solution to a numericaloptimization problem rather than an explicit modeling assumption. Bayes-optimalagents are risk-neutral, since they solely attune to the expected return, andambiguity-neutral, since they act in new situations as if the uncertainty wereknown. This is in contrast to risk-sensitive agents, which additionally exploitthe higher-order moments of the return, and ambiguity-sensitive agents, whichact differently when recognizing situations in which they lack knowledge.Humans are also known to be averse to ambiguity and sensitive to risk in waysthat aren't Bayes-optimal, indicating that such sensitivity can conferadvantages, especially in safety-critical situations. How can we extend themeta-learning protocol to generate risk- and ambiguity-sensitive agents? Thegoal of this work is to fill this gap in the literature by showing that risk-and ambiguity-sensitivity also emerge as the result of an optimization problemusing modified meta-training algorithms, which manipulate theexperience-generation process of the learner. We empirically test our proposedmeta-training algorithms on agents exposed to foundational classes ofdecision-making experiments and demonstrate that they become sensitive to riskand ambiguity.

Beyond Bayes-optimality: meta-learning what you know you don't know