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

Recent developments in machine-learning algorithms have led to impressiveperformance increases in many traditional application scenarios of artificialintelligence research. In the area of deep reinforcement learning, deeplearning functional architectures are combined with incremental learningschemes for sequential tasks that include interaction-based, but often delayedfeedback. Despite their impressive successes, modern machine-learningapproaches, including deep reinforcement learning, still perform weakly whencompared to flexibly adaptive biological systems in certain naturally occurringscenarios. Such scenarios include transfers to environments different than theones in which the training took place or environments that dynamically change,both of which are often mastered by biological systems through a capabilitythat we here term "fluid adaptivity" to contrast it from the much sloweradaptivity ("crystallized adaptivity") of the prior learning from which thebehavior emerged. In this article, we derive and discuss research strategies,based on analyzes of fluid adaptivity in biological systems and its neuronalmodeling, that might aid in equipping future artificially intelligent systemswith capabilities of fluid adaptivity more similar to those seen in somebiologically intelligent systems. A key component of this research strategy isthe dynamization of the problem space itself and the implementation of thisdynamization by suitably designed flexibly interacting modules.

From Crystallized Adaptivity to Fluid Adaptivity in Deep Reinforcement Learning -- Insights from Biological Systems on Adaptive Flexibility