FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2899138465> ?p ?o ?g. }

• W2899138465 abstract "Humans have a remarkable ability to learn new concepts from only a few examples and quickly adapt to unforeseen circumstances. To do so, they build upon their prior experience and prepare for the ability to adapt, allowing the combination of previous observations with small amounts of new evidence for fast learning. In most machine learning systems, however, there are distinct train and test phases: training consists of updating the model using data, and at test time, the model is deployed as a rigid decision-making engine. In this thesis, we discuss gradient-based algorithms for learning to learn, or meta-learning, which aim to endow machines with flexibility akin to that of humans. Instead of deploying a fixed, non-adaptable system, these meta-learning techniques explicitly train for the ability to quickly adapt so that, at test time, they can learn quickly when faced with new scenarios.To study the problem of learning to learn, we first develop a clear and formal definition of the meta-learning problem, its terminology, and desirable properties of meta-learning algorithms. Building upon these foundations, we present a class of model-agnostic meta-learning methods that embed gradient-based optimization into the learner. Unlike prior approaches to learning to learn, this class of methods focus on acquiring a transferable representation rather than a good learning rule. As a result, these methods inherit a number of desirable properties from using a fixed optimization as the learning rule, while still maintaining full expressivity, since the learned representations can control the update rule.We show how these methods can be extended for applications in motor control by combining elements of meta-learning with techniques for deep model-based reinforcement learning, imitation learning, and inverse reinforcement learning. By doing so, we build simulated agents that can adapt in dynamic environments, enable real robots to learn to manipulate new objects by watching a video of a human, and allow humans to convey goals to robots with only a few images. Finally, we conclude by discussing open questions and future directions in meta-learning, aiming to identify the key shortcomings and limiting assumptions of our existing approaches." @default.
• W2899138465 created "2018-11-09" @default.
• W2899138465 creator A5005431772 @default.
• W2899138465 date "2018-01-01" @default.
• W2899138465 modified "2023-09-25" @default.
• W2899138465 title "Learning to Learn with Gradients" @default.
• W2899138465 cites W1500462030 @default.
• W2899138465 cites W1516311991 @default.
• W2899138465 cites W1567512734 @default.
• W2899138465 cites W1591675293 @default.
• W2899138465 cites W1598377843 @default.
• W2899138465 cites W1757796397 @default.
• W2899138465 cites W1777239053 @default.
• W2899138465 cites W1868018859 @default.
• W2899138465 cites W1903029394 @default.
• W2899138465 cites W1929981607 @default.
• W2899138465 cites W195033972 @default.
• W2899138465 cites W1968962398 @default.
• W2899138465 cites W1969074599 @default.
• W2899138465 cites W1981494355 @default.
• W2899138465 cites W1988348003 @default.
• W2899138465 cites W1994648061 @default.
• W2899138465 cites W1998534269 @default.
• W2899138465 cites W1999874108 @default.
• W2899138465 cites W2012204020 @default.
• W2899138465 cites W2012585528 @default.
• W2899138465 cites W2016765487 @default.
• W2899138465 cites W2030290736 @default.
• W2899138465 cites W2061562262 @default.
• W2899138465 cites W2062179223 @default.
• W2899138465 cites W2074692319 @default.
• W2899138465 cites W2081034428 @default.
• W2899138465 cites W2104068492 @default.
• W2899138465 cites W2104733512 @default.
• W2899138465 cites W2108677974 @default.
• W2899138465 cites W2115413618 @default.
• W2899138465 cites W2117675763 @default.
• W2899138465 cites W2121103318 @default.
• W2899138465 cites W2122136962 @default.
• W2899138465 cites W2124695578 @default.
• W2899138465 cites W2125612430 @default.
• W2899138465 cites W2125930537 @default.
• W2899138465 cites W2130726249 @default.
• W2899138465 cites W2133068870 @default.
• W2899138465 cites W2137310197 @default.
• W2899138465 cites W2137825550 @default.
• W2899138465 cites W2140804329 @default.
• W2899138465 cites W2148522164 @default.
• W2899138465 cites W2149933564 @default.
• W2899138465 cites W2155541015 @default.
• W2899138465 cites W2158782408 @default.
• W2899138465 cites W2158815628 @default.
• W2899138465 cites W2160609165 @default.
• W2899138465 cites W2161395589 @default.
• W2899138465 cites W2162708558 @default.
• W2899138465 cites W2169498096 @default.
• W2899138465 cites W2170973209 @default.
• W2899138465 cites W2173248099 @default.
• W2899138465 cites W2174786457 @default.
• W2899138465 cites W2201912979 @default.
• W2899138465 cites W2287334441 @default.
• W2899138465 cites W2290354866 @default.
• W2899138465 cites W2342662072 @default.
• W2899138465 cites W2401823607 @default.
• W2899138465 cites W2409942531 @default.
• W2899138465 cites W2412589713 @default.
• W2899138465 cites W2472819217 @default.
• W2899138465 cites W2513173501 @default.
• W2899138465 cites W2519882289 @default.
• W2899138465 cites W2523013761 @default.
• W2899138465 cites W2525778437 @default.
• W2899138465 cites W2528489519 @default.
• W2899138465 cites W2529658650 @default.
• W2899138465 cites W2544683879 @default.
• W2899138465 cites W2584009249 @default.
• W2899138465 cites W2591957724 @default.
• W2899138465 cites W2601593996 @default.
• W2899138465 cites W2604763608 @default.
• W2899138465 cites W2605368761 @default.
• W2899138465 cites W2753160622 @default.
• W2899138465 cites W2762872434 @default.
• W2899138465 cites W2766363782 @default.
• W2899138465 cites W2769555377 @default.
• W2899138465 cites W2787501667 @default.
• W2899138465 cites W2949117887 @default.
• W2899138465 cites W2951458352 @default.
• W2899138465 cites W2953044442 @default.
• W2899138465 cites W2962690307 @default.
• W2899138465 cites W2962695963 @default.
• W2899138465 cites W2962787403 @default.
• W2899138465 cites W2962839807 @default.
• W2899138465 cites W2962871243 @default.
• W2899138465 cites W2962887844 @default.
• W2899138465 cites W2962951365 @default.
• W2899138465 cites W2963207607 @default.
• W2899138465 cites W2963277051 @default.
• W2899138465 cites W2963280855 @default.
• W2899138465 cites W2963341924 @default.
• W2899138465 cites W2963430173 @default.
• W2899138465 cites W2963577640 @default.

• next