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/

Zendy Plus

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

Zendy Tools

Zendy Open

While deep reinforcement learning techniques have led to agents that aresuccessfully able to learn to perform a number of tasks that had beenpreviously unlearnable, these techniques are still susceptible to thelongstanding problem of {\em reward sparsity}. This is especially true fortasks such as training an agent to play StarCraft II, a real-time strategy gamewhere reward is only given at the end of a game which is usually very long.While this problem can be addressed through reward shaping, such approachestypically require a human expert with specialized knowledge. Inspired by thevision of enabling reward shaping through the more-accessible paradigm ofnatural-language narration, we investigate to what extent we can contextualizethese narrations by grounding them to the goal-specific states. We present amutual-embedding model using a multi-input deep-neural network that projects asequence of natural language commands into the same high-dimensionalrepresentation space as corresponding goal states. We show that using thismodel we can learn an embedding space with separable and distinct clusters thataccurately maps natural-language commands to corresponding game states . Wealso discuss how this model can allow for the use of narrations as a robustform of reward shaping to improve RL performance and efficiency.

Grounding Natural Language Commands to StarCraft II Game States for Narration-Guided Reinforcement Learning