Descrizione del progetto
L’apprendimento per rinforzo fa un grande salto in avanti
Imparare dall’interazione è un’idea fondamentale alla base di quasi tutte le teorie dell’apprendimento e dell’intelligenza. Rispetto ad altri approcci all’apprendimento automatico, l’apprendimento per rinforzo si concentra molto di più sulle questioni computazionali che sorgono quando gli agenti del software imparano dalle interazioni con un ambiente. Il progetto COLT-MDP, finanziato dall’UE, si propone di far progredire gli approcci teorici più avanzati per rafforzare l’apprendimento. Lo studio ruoterà intorno a tre pilastri. La progettazione di modelli di rappresentazione compatta permetterà ai ricercatori di strutturare meglio problemi complessi. Ciò potrebbe estendere l’applicabilità dell’apprendimento per rinforzo a molti campi. Inoltre, i ricercatori svilupperanno algoritmi di calcolo efficienti. Sotto la spinta delle preoccupazioni dell’opinione pubblica, un crescente corpus di ricerche affronterà le questioni relative all’equità e alla privacy nell’intelligenza artificiale.
Obiettivo
Computational learning theory has been highly successful over the last three decades, both in providing deep mathematical theories and in influencing the practice of machine learning. One of the great recent successes of computational learning theory has been the study of online learning and multi-arm bandits. This line of research has been highly successful, both theoretically and practically, addressing many important applications. Unfortunately, the recent theoretical progress in Markov Decision Process and reinforcement learning has been slower.
Based on my fundamental contributions to reinforcement learning (e.g. policy gradient, sparse sampling and trajectory trees), to online learning and machine learning in general, I propose to take the theoretical and practical success of online learning to the “next level” by making significant breakthroughs in reinforcement learning. Our main aim is to advance the state of the art in the theory of reinforcement learning, and our research will revolve around three pillars: (1) compact representation, (2) efficient computation and (3) societal challenges, including fairness and privacy.
A successful project will greatly impact reinforcement learning in all its stages. Modelling: Introducing new compact representation models, will enhance our understanding how to structure complex problems, which would greatly extend the applicability of reinforcement learning. Efficient computation: New algorithmic methodologies will give new insight for overcoming computational and statistical barriers both for planning and learning. Learning: New learning paradigms would address fundamental issues of copping with uncertainties in complex control environments of reinforcement learning. Societal challenges: Allowing the community to understand, assess, address and overcome societal challenges is of the greatest importance to the acceptance of AI methodologies by the general public.
Campo scientifico
Programma(i)
Argomento(i)
Meccanismo di finanziamento
ERC-ADG - Advanced GrantIstituzione ospitante
69978 Tel Aviv
Israele