AI brief
Apple Machine Learning Research describes SCLATE, a substrate for training and evaluating continual-learning agents. The system addresses agents that operate as models, harnesses and memory across long multi-session horizons, interleaving tasks with agent-side events such as session stop and start, crons, and memory consolidation.
Why it matters: It targets infrastructure for training and evaluating agents that work over long, multi-session periods rather than single tasks.
Written by AI from Apple Machine Learning Research's published text.
Source
“Continual-learning agents are systems of models, harnesses, and memory operating over long multi-session horizons. Evaluating and training them requires interleaving tasks with agent-side events such as session stop and start, crons, and memory consolidation.”
A paper, laboratory publication or research report.
Read original story ↗