Matches in SemOpenAlex for { <https://semopenalex.org/work/W2743553019> ?p ?o ?g. }
- W2743553019 abstract "We tackle the problem of learning robotic sensorimotor control policies that can generalize to visually diverse and unseen environments. Achieving broad generalization typically requires large datasets, which are difficult to obtain for task-specific interactive processes such as reinforcement learning or learning from demonstration. However, much of the visual diversity in the world can be captured through passively collected datasets of images or videos. In our method, which we refer to as GPLAC (Generalized Policy Learning with Attentional Classifier), we use both interaction data and weakly labeled image data to augment the generalization capacity of sensorimotor policies. Our method combines multitask learning on action selection and an auxiliary binary classification objective, together with a convolutional neural network architecture that uses an attentional mechanism to avoid distractors. We show that pairing interaction data from just a single environment with a diverse dataset of weakly labeled data results in greatly improved generalization to unseen environments, and show that this generalization depends on both the auxiliary objective and the attentional architecture that we propose. We demonstrate our results in both simulation and on a real robotic manipulator, and demonstrate substantial improvement over standard convolutional architectures and domain adaptation methods." @default.
- W2743553019 created "2017-08-17" @default.
- W2743553019 creator A5026322200 @default.
- W2743553019 creator A5071640510 @default.
- W2743553019 creator A5078558551 @default.
- W2743553019 date "2017-08-07" @default.
- W2743553019 modified "2023-09-27" @default.
- W2743553019 title "GPLAC: Generalizing Vision-Based Robotic Skills using Weakly Labeled Images" @default.
- W2743553019 cites W1503925285 @default.
- W2743553019 cites W1514535095 @default.
- W2743553019 cites W1522301498 @default.
- W2743553019 cites W1564897360 @default.
- W2743553019 cites W1577269164 @default.
- W2743553019 cites W1686810756 @default.
- W2743553019 cites W1850742715 @default.
- W2743553019 cites W1882958252 @default.
- W2743553019 cites W1896424170 @default.
- W2743553019 cites W1932624639 @default.
- W2743553019 cites W1989021449 @default.
- W2743553019 cites W1994488211 @default.
- W2743553019 cites W2038794597 @default.
- W2743553019 cites W2103104224 @default.
- W2743553019 cites W2108766544 @default.
- W2743553019 cites W2109447584 @default.
- W2743553019 cites W2112474089 @default.
- W2743553019 cites W2112796928 @default.
- W2743553019 cites W2124172617 @default.
- W2743553019 cites W2148461049 @default.
- W2743553019 cites W2158782408 @default.
- W2743553019 cites W2163605009 @default.
- W2743553019 cites W2167224731 @default.
- W2743553019 cites W2171810632 @default.
- W2743553019 cites W2173393671 @default.
- W2743553019 cites W2210483910 @default.
- W2743553019 cites W2280377497 @default.
- W2743553019 cites W2290564286 @default.
- W2743553019 cites W2293467699 @default.
- W2743553019 cites W2342840547 @default.
- W2743553019 cites W2432004435 @default.
- W2743553019 cites W2473208550 @default.
- W2743553019 cites W2528973876 @default.
- W2743553019 cites W259216465 @default.
- W2743553019 cites W2949098821 @default.
- W2743553019 cites W2949416428 @default.
- W2743553019 cites W2949608212 @default.
- W2743553019 cites W2950612966 @default.
- W2743553019 cites W2951004968 @default.
- W2743553019 cites W2951326654 @default.
- W2743553019 cites W2951670162 @default.
- W2743553019 cites W2951751411 @default.
- W2743553019 cites W2952229419 @default.
- W2743553019 cites W2952623155 @default.
- W2743553019 cites W2952629144 @default.
- W2743553019 cites W2953127297 @default.
- W2743553019 cites W2963430173 @default.
- W2743553019 cites W2963784072 @default.
- W2743553019 cites W2964161785 @default.
- W2743553019 cites W630242894 @default.
- W2743553019 hasPublicationYear "2017" @default.
- W2743553019 type Work @default.
- W2743553019 sameAs 2743553019 @default.
- W2743553019 citedByCount "3" @default.
- W2743553019 countsByYear W27435530192017 @default.
- W2743553019 countsByYear W27435530192019 @default.
- W2743553019 countsByYear W27435530192020 @default.
- W2743553019 crossrefType "posted-content" @default.
- W2743553019 hasAuthorship W2743553019A5026322200 @default.
- W2743553019 hasAuthorship W2743553019A5071640510 @default.
- W2743553019 hasAuthorship W2743553019A5078558551 @default.
- W2743553019 hasConcept C119857082 @default.
- W2743553019 hasConcept C120665830 @default.
- W2743553019 hasConcept C121332964 @default.
- W2743553019 hasConcept C12267149 @default.
- W2743553019 hasConcept C134306372 @default.
- W2743553019 hasConcept C139807058 @default.
- W2743553019 hasConcept C154945302 @default.
- W2743553019 hasConcept C162324750 @default.
- W2743553019 hasConcept C177148314 @default.
- W2743553019 hasConcept C187736073 @default.
- W2743553019 hasConcept C2776434776 @default.
- W2743553019 hasConcept C2780451532 @default.
- W2743553019 hasConcept C33923547 @default.
- W2743553019 hasConcept C41008148 @default.
- W2743553019 hasConcept C48372109 @default.
- W2743553019 hasConcept C66905080 @default.
- W2743553019 hasConcept C81363708 @default.
- W2743553019 hasConcept C94375191 @default.
- W2743553019 hasConcept C95623464 @default.
- W2743553019 hasConcept C97541855 @default.
- W2743553019 hasConceptScore W2743553019C119857082 @default.
- W2743553019 hasConceptScore W2743553019C120665830 @default.
- W2743553019 hasConceptScore W2743553019C121332964 @default.
- W2743553019 hasConceptScore W2743553019C12267149 @default.
- W2743553019 hasConceptScore W2743553019C134306372 @default.
- W2743553019 hasConceptScore W2743553019C139807058 @default.
- W2743553019 hasConceptScore W2743553019C154945302 @default.
- W2743553019 hasConceptScore W2743553019C162324750 @default.
- W2743553019 hasConceptScore W2743553019C177148314 @default.
- W2743553019 hasConceptScore W2743553019C187736073 @default.
- W2743553019 hasConceptScore W2743553019C2776434776 @default.