Matches in SemOpenAlex for { <https://semopenalex.org/work/W3114605215> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W3114605215 abstract "In previous work, using a process we call meshing, the reachable state spaces for various continuous and hybrid systems were approximated as a discrete set of states which can then be synthesized into a Markov chain. One of the applications for this approach has been to analyze locomotion policies obtained by reinforcement learning, in a step towards making empirical guarantees about the stability properties of the resulting system. In a separate line of research, we introduced a modified reward function for on-policy reinforcement learning algorithms that utilizes a of rollout trajectories. This reward was shown to encourage policies that induce individual trajectories which can be more compactly represented as a discrete mesh. In this work we combine these two threads of research by building meshes of the reachable state space of a system subject to disturbances and controlled by policies obtained with the modified reward. Our analysis shows that the modified policies do produce much smaller reachable meshes. This shows that agents trained with the fractal dimension reward transfer their desirable quality of having a more compact state space to a setting with external disturbances. The results also suggest that the previous work using mesh based tools to analyze RL policies may be extended to higher dimensional systems or to higher resolution meshes than would have otherwise been possible." @default.
- W3114605215 created "2021-01-05" @default.
- W3114605215 creator A5038111090 @default.
- W3114605215 creator A5071708309 @default.
- W3114605215 date "2020-12-24" @default.
- W3114605215 modified "2023-09-27" @default.
- W3114605215 title "Mesh Based Analysis of Low Fractal Dimension ReinforcementLearning Policies" @default.
- W3114605215 cites W1982880998 @default.
- W3114605215 cites W2294535740 @default.
- W3114605215 cites W2726187156 @default.
- W3114605215 cites W2735039421 @default.
- W3114605215 cites W2885163910 @default.
- W3114605215 cites W2890326782 @default.
- W3114605215 cites W2911087563 @default.
- W3114605215 cites W2913754617 @default.
- W3114605215 cites W2929842775 @default.
- W3114605215 cites W2972420235 @default.
- W3114605215 cites W3210235935 @default.
- W3114605215 hasPublicationYear "2020" @default.
- W3114605215 type Work @default.
- W3114605215 sameAs 3114605215 @default.
- W3114605215 citedByCount "0" @default.
- W3114605215 crossrefType "posted-content" @default.
- W3114605215 hasAuthorship W3114605215A5038111090 @default.
- W3114605215 hasAuthorship W3114605215A5071708309 @default.
- W3114605215 hasConcept C105795698 @default.
- W3114605215 hasConcept C111919701 @default.
- W3114605215 hasConcept C112972136 @default.
- W3114605215 hasConcept C11413529 @default.
- W3114605215 hasConcept C119857082 @default.
- W3114605215 hasConcept C121684516 @default.
- W3114605215 hasConcept C126255220 @default.
- W3114605215 hasConcept C154945302 @default.
- W3114605215 hasConcept C159886148 @default.
- W3114605215 hasConcept C177264268 @default.
- W3114605215 hasConcept C199360897 @default.
- W3114605215 hasConcept C202444582 @default.
- W3114605215 hasConcept C2778572836 @default.
- W3114605215 hasConcept C31487907 @default.
- W3114605215 hasConcept C33676613 @default.
- W3114605215 hasConcept C33923547 @default.
- W3114605215 hasConcept C41008148 @default.
- W3114605215 hasConcept C72434380 @default.
- W3114605215 hasConcept C97541855 @default.
- W3114605215 hasConcept C98763669 @default.
- W3114605215 hasConceptScore W3114605215C105795698 @default.
- W3114605215 hasConceptScore W3114605215C111919701 @default.
- W3114605215 hasConceptScore W3114605215C112972136 @default.
- W3114605215 hasConceptScore W3114605215C11413529 @default.
- W3114605215 hasConceptScore W3114605215C119857082 @default.
- W3114605215 hasConceptScore W3114605215C121684516 @default.
- W3114605215 hasConceptScore W3114605215C126255220 @default.
- W3114605215 hasConceptScore W3114605215C154945302 @default.
- W3114605215 hasConceptScore W3114605215C159886148 @default.
- W3114605215 hasConceptScore W3114605215C177264268 @default.
- W3114605215 hasConceptScore W3114605215C199360897 @default.
- W3114605215 hasConceptScore W3114605215C202444582 @default.
- W3114605215 hasConceptScore W3114605215C2778572836 @default.
- W3114605215 hasConceptScore W3114605215C31487907 @default.
- W3114605215 hasConceptScore W3114605215C33676613 @default.
- W3114605215 hasConceptScore W3114605215C33923547 @default.
- W3114605215 hasConceptScore W3114605215C41008148 @default.
- W3114605215 hasConceptScore W3114605215C72434380 @default.
- W3114605215 hasConceptScore W3114605215C97541855 @default.
- W3114605215 hasConceptScore W3114605215C98763669 @default.
- W3114605215 hasLocation W31146052151 @default.
- W3114605215 hasOpenAccess W3114605215 @default.
- W3114605215 hasPrimaryLocation W31146052151 @default.
- W3114605215 hasRelatedWork W103181800 @default.
- W3114605215 hasRelatedWork W1040735863 @default.
- W3114605215 hasRelatedWork W1570523567 @default.
- W3114605215 hasRelatedWork W162145636 @default.
- W3114605215 hasRelatedWork W2198325613 @default.
- W3114605215 hasRelatedWork W23560277 @default.
- W3114605215 hasRelatedWork W2739211902 @default.
- W3114605215 hasRelatedWork W2911722107 @default.
- W3114605215 hasRelatedWork W2951029270 @default.
- W3114605215 hasRelatedWork W2958708366 @default.
- W3114605215 hasRelatedWork W2979435954 @default.
- W3114605215 hasRelatedWork W3010257396 @default.
- W3114605215 hasRelatedWork W3014710427 @default.
- W3114605215 hasRelatedWork W3024764221 @default.
- W3114605215 hasRelatedWork W3048304628 @default.
- W3114605215 hasRelatedWork W3110729291 @default.
- W3114605215 hasRelatedWork W3116610253 @default.
- W3114605215 hasRelatedWork W3202124232 @default.
- W3114605215 hasRelatedWork W3212368910 @default.
- W3114605215 hasRelatedWork W3099303842 @default.
- W3114605215 isParatext "false" @default.
- W3114605215 isRetracted "false" @default.
- W3114605215 magId "3114605215" @default.
- W3114605215 workType "article" @default.