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

Intelligent agents need to remember salient information to reason inpartially-observed environments. For example, agents with a first-person viewshould remember the positions of relevant objects even if they go out of view.Similarly, to effectively navigate through rooms agents need to remember thefloor plan of how rooms are connected. However, most benchmark tasks inreinforcement learning do not test long-term memory in agents, slowing downprogress in this important research direction. In this paper, we introduce theMemory Maze, a 3D domain of randomized mazes specifically designed forevaluating long-term memory in agents. Unlike existing benchmarks, Memory Mazemeasures long-term memory separate from confounding agent abilities andrequires the agent to localize itself by integrating information over time.With Memory Maze, we propose an online reinforcement learning benchmark, adiverse offline dataset, and an offline probing evaluation. Recording a humanplayer establishes a strong baseline and verifies the need to build up andretain memories, which is reflected in their gradually increasing rewardswithin each episode. We find that current algorithms benefit from training withtruncated backpropagation through time and succeed on small mazes, but fallshort of human performance on the large mazes, leaving room for futurealgorithmic designs to be evaluated on the Memory Maze.

Evaluating Long-Term Memory in 3D Mazes