FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2763393077> ?p ?o ?g. }

• W2763393077 abstract "Abstract Humans are remarkably adept at generalizing knowledge between experiences in a way that can be difficult for computers. Often, this entails generalizing constituent pieces of experiences that do not fully overlap, but nonetheless share useful similarities with, previously acquired knowledge. However, it is often unclear how knowledge gained in one context should generalize to another. Previous computational models and data suggest that rather than learning about each individual context, humans build latent abstract structures and learn to link these structures to arbitrary contexts, facilitating generalization. In these models, task structures that are more popular across contexts are more likely to be revisited in new contexts. However, these models can only re-use policies as a whole and are unable to transfer knowledge about the transition structure of the environment even if only the goal has changed (or vice-versa). This contrasts with ecological settings, where some aspects of task structure, such as the transition function, will be shared between context separately from other aspects, such as the reward function. Here, we develop a novel non-parametric Bayesian agent that forms independent latent clusters for transition and reward functions, affording separable transfer of their constituent parts across contexts. We show that the relative performance of this agent compared to an agent that jointly clusters reward and transition functions depends environmental task statistics: the mutual information between transition and reward functions and the stochasticity of the observations. We formalize our analysis through an information theoretic account of the priors, and propose a meta learning agent that dynamically arbitrates between strategies across task domains to optimize a statistical tradeoff. Author summary A musician may learn to generalize behaviors across instruments for different purposes, for example, reusing hand motions used when playing classical on the flute to play jazz on the saxophone. Conversely, she may learn to play a single song across many instruments that require completely distinct physical motions, but nonetheless transfer knowledge between them. This degree of compositionality is often absent from computational frameworks of learning, forcing agents either to generalize entire learned policies or to learn new policies from scratch. Here, we propose a solution to this problem that allows an agent to generalize components of a policy independently and compare it to an agent that generalizes components as a whole. We show that the degree to which one form of generalization is favored over the other is dependent on the features of task domain, with independent generalization of task components favored in environments with weak relationships between components or high degrees of noise and joint generalization of task components favored when there is a clear, discoverable relationship between task components. Furthermore, we show that the overall meta structure of the environment can be learned and leveraged by an agent that dynamically arbitrates between these forms of structure learning." @default.
• W2763393077 created "2017-10-20" @default.
• W2763393077 creator A5007609257 @default.
• W2763393077 creator A5064851462 @default.
• W2763393077 date "2017-10-02" @default.
• W2763393077 modified "2023-09-28" @default.
• W2763393077 title "Compositional clustering in task structure learning" @default.
• W2763393077 cites W1988520084 @default.
• W2763393077 cites W2006114123 @default.
• W2763393077 cites W2032416253 @default.
• W2763393077 cites W2038407704 @default.
• W2763393077 cites W2039522160 @default.
• W2763393077 cites W2059569317 @default.
• W2763393077 cites W2061304498 @default.
• W2763393077 cites W2072916238 @default.
• W2763393077 cites W2085679424 @default.
• W2763393077 cites W2086710210 @default.
• W2763393077 cites W2101355568 @default.
• W2763393077 cites W2107139863 @default.
• W2763393077 cites W2107726111 @default.
• W2763393077 cites W2109910161 @default.
• W2763393077 cites W2113122939 @default.
• W2763393077 cites W2118373646 @default.
• W2763393077 cites W2136450052 @default.
• W2763393077 cites W2140956413 @default.
• W2763393077 cites W2149565728 @default.
• W2763393077 cites W2167956961 @default.
• W2763393077 cites W2168342951 @default.
• W2763393077 cites W2276275925 @default.
• W2763393077 cites W2341634245 @default.
• W2763393077 cites W2424475825 @default.
• W2763393077 cites W2517275106 @default.
• W2763393077 cites W2542993203 @default.
• W2763393077 cites W263845233 @default.
• W2763393077 cites W2782189646 @default.
• W2763393077 cites W2951066214 @default.
• W2763393077 cites W2953319434 @default.
• W2763393077 cites W2962872206 @default.
• W2763393077 cites W2963305465 @default.
• W2763393077 cites W778742492 @default.
• W2763393077 doi "https://doi.org/10.1101/196923" @default.
• W2763393077 hasPublicationYear "2017" @default.
• W2763393077 type Work @default.
• W2763393077 sameAs 2763393077 @default.
• W2763393077 citedByCount "0" @default.
• W2763393077 crossrefType "posted-content" @default.
• W2763393077 hasAuthorship W2763393077A5007609257 @default.
• W2763393077 hasAuthorship W2763393077A5064851462 @default.
• W2763393077 hasBestOaLocation W27633930771 @default.
• W2763393077 hasConcept C104317684 @default.
• W2763393077 hasConcept C107673813 @default.
• W2763393077 hasConcept C119857082 @default.
• W2763393077 hasConcept C134306372 @default.
• W2763393077 hasConcept C14036430 @default.
• W2763393077 hasConcept C151730666 @default.
• W2763393077 hasConcept C154945302 @default.
• W2763393077 hasConcept C162324750 @default.
• W2763393077 hasConcept C177148314 @default.
• W2763393077 hasConcept C185592680 @default.
• W2763393077 hasConcept C187736073 @default.
• W2763393077 hasConcept C194232998 @default.
• W2763393077 hasConcept C2779343474 @default.
• W2763393077 hasConcept C2780451532 @default.
• W2763393077 hasConcept C33923547 @default.
• W2763393077 hasConcept C41008148 @default.
• W2763393077 hasConcept C55493867 @default.
• W2763393077 hasConcept C73555534 @default.
• W2763393077 hasConcept C78458016 @default.
• W2763393077 hasConcept C86803240 @default.
• W2763393077 hasConceptScore W2763393077C104317684 @default.
• W2763393077 hasConceptScore W2763393077C107673813 @default.
• W2763393077 hasConceptScore W2763393077C119857082 @default.
• W2763393077 hasConceptScore W2763393077C134306372 @default.
• W2763393077 hasConceptScore W2763393077C14036430 @default.
• W2763393077 hasConceptScore W2763393077C151730666 @default.
• W2763393077 hasConceptScore W2763393077C154945302 @default.
• W2763393077 hasConceptScore W2763393077C162324750 @default.
• W2763393077 hasConceptScore W2763393077C177148314 @default.
• W2763393077 hasConceptScore W2763393077C185592680 @default.
• W2763393077 hasConceptScore W2763393077C187736073 @default.
• W2763393077 hasConceptScore W2763393077C194232998 @default.
• W2763393077 hasConceptScore W2763393077C2779343474 @default.
• W2763393077 hasConceptScore W2763393077C2780451532 @default.
• W2763393077 hasConceptScore W2763393077C33923547 @default.
• W2763393077 hasConceptScore W2763393077C41008148 @default.
• W2763393077 hasConceptScore W2763393077C55493867 @default.
• W2763393077 hasConceptScore W2763393077C73555534 @default.
• W2763393077 hasConceptScore W2763393077C78458016 @default.
• W2763393077 hasConceptScore W2763393077C86803240 @default.
• W2763393077 hasLocation W27633930771 @default.
• W2763393077 hasLocation W27633930772 @default.
• W2763393077 hasLocation W27633930773 @default.
• W2763393077 hasLocation W27633930774 @default.
• W2763393077 hasOpenAccess W2763393077 @default.
• W2763393077 hasPrimaryLocation W27633930771 @default.
• W2763393077 hasRelatedWork W1123419 @default.
• W2763393077 hasRelatedWork W2181750757 @default.
• W2763393077 hasRelatedWork W2266522507 @default.
• W2763393077 hasRelatedWork W2295582178 @default.
• W2763393077 hasRelatedWork W2402997934 @default.

• next