Matches in SemOpenAlex for { <https://semopenalex.org/work/W197688093> ?p ?o ?g. }
Showing items 1 to 78 of
78
with 100 items per page.
- W197688093 abstract "Reinforcement learning treats each input, feature, or stimulus as having a positive or negative reward value. Some stimuli, however, negate or inhibit the values of certain other predictors (excitors) when presented with them, but are otherwise neutral. We show that both linear and non-linear value-function approximators assign inhibitory features a strong value with the opposite valence of the predictor it inhibits (i.e., inhibitor= -excitor). In one circumstance, this gives a correct prediction (i.e., excitor + inhibitor = neutral outcome). Importantly, however, value-function approximators incorrectly predict that when the inhibitor is presented alone, a negative or oppositely valenced outcome will follow whereas the inhibitor alone is actually followed by a neutral outcome. Essentially, we show that having reward value as a direct predictive target can make inhibitors indistinguishable from excitors that predict the oppositely valenced outcome. We show that this problem can be easily avoided if the reinforcement learning problem is broken into 1) a supervised learning module that predicts the positive appearance of primary reinforcements and 2) a reinforcement learning module which sums their agent-defined values." @default.
- W197688093 created "2016-06-24" @default.
- W197688093 creator A5014253093 @default.
- W197688093 creator A5088920905 @default.
- W197688093 date "2013-12-20" @default.
- W197688093 modified "2023-09-26" @default.
- W197688093 title "An Architecture for Distinguishing between Predictors and Inhibitors in Reinforcement Learning" @default.
- W197688093 hasPublicationYear "2013" @default.
- W197688093 type Work @default.
- W197688093 sameAs 197688093 @default.
- W197688093 citedByCount "0" @default.
- W197688093 crossrefType "proceedings-article" @default.
- W197688093 hasAuthorship W197688093A5014253093 @default.
- W197688093 hasAuthorship W197688093A5088920905 @default.
- W197688093 hasConcept C119857082 @default.
- W197688093 hasConcept C121332964 @default.
- W197688093 hasConcept C126322002 @default.
- W197688093 hasConcept C144237770 @default.
- W197688093 hasConcept C148220186 @default.
- W197688093 hasConcept C154945302 @default.
- W197688093 hasConcept C15744967 @default.
- W197688093 hasConcept C168900304 @default.
- W197688093 hasConcept C180747234 @default.
- W197688093 hasConcept C2776291640 @default.
- W197688093 hasConcept C2779918689 @default.
- W197688093 hasConcept C3019719930 @default.
- W197688093 hasConcept C33923547 @default.
- W197688093 hasConcept C41008148 @default.
- W197688093 hasConcept C62520636 @default.
- W197688093 hasConcept C67203356 @default.
- W197688093 hasConcept C71924100 @default.
- W197688093 hasConcept C77805123 @default.
- W197688093 hasConcept C97541855 @default.
- W197688093 hasConceptScore W197688093C119857082 @default.
- W197688093 hasConceptScore W197688093C121332964 @default.
- W197688093 hasConceptScore W197688093C126322002 @default.
- W197688093 hasConceptScore W197688093C144237770 @default.
- W197688093 hasConceptScore W197688093C148220186 @default.
- W197688093 hasConceptScore W197688093C154945302 @default.
- W197688093 hasConceptScore W197688093C15744967 @default.
- W197688093 hasConceptScore W197688093C168900304 @default.
- W197688093 hasConceptScore W197688093C180747234 @default.
- W197688093 hasConceptScore W197688093C2776291640 @default.
- W197688093 hasConceptScore W197688093C2779918689 @default.
- W197688093 hasConceptScore W197688093C3019719930 @default.
- W197688093 hasConceptScore W197688093C33923547 @default.
- W197688093 hasConceptScore W197688093C41008148 @default.
- W197688093 hasConceptScore W197688093C62520636 @default.
- W197688093 hasConceptScore W197688093C67203356 @default.
- W197688093 hasConceptScore W197688093C71924100 @default.
- W197688093 hasConceptScore W197688093C77805123 @default.
- W197688093 hasConceptScore W197688093C97541855 @default.
- W197688093 hasLocation W1976880931 @default.
- W197688093 hasOpenAccess W197688093 @default.
- W197688093 hasPrimaryLocation W1976880931 @default.
- W197688093 hasRelatedWork W1870203926 @default.
- W197688093 hasRelatedWork W1992373090 @default.
- W197688093 hasRelatedWork W2023476462 @default.
- W197688093 hasRelatedWork W2072996399 @default.
- W197688093 hasRelatedWork W2080782270 @default.
- W197688093 hasRelatedWork W2084822470 @default.
- W197688093 hasRelatedWork W2245629054 @default.
- W197688093 hasRelatedWork W2315488191 @default.
- W197688093 hasRelatedWork W2316485682 @default.
- W197688093 hasRelatedWork W2408481333 @default.
- W197688093 hasRelatedWork W2591789459 @default.
- W197688093 hasRelatedWork W2612765367 @default.
- W197688093 hasRelatedWork W2767011876 @default.
- W197688093 hasRelatedWork W2888939091 @default.
- W197688093 hasRelatedWork W2895764062 @default.
- W197688093 hasRelatedWork W2952283715 @default.
- W197688093 hasRelatedWork W3014457489 @default.
- W197688093 hasRelatedWork W3125893104 @default.
- W197688093 hasRelatedWork W405892556 @default.
- W197688093 isParatext "false" @default.
- W197688093 isRetracted "false" @default.
- W197688093 magId "197688093" @default.
- W197688093 workType "article" @default.