The purpose of this technical report is two-fold. First of all, it introduces a suite of challenging continuous control tasks (integrated with OpenAI Gym) based on currently existing robotics hardware. The tasks include pushing, sliding and pick & place with a Fetch robotic arm as well as in-hand object manipulation with a Shadow Dexterous Hand. All tasks have sparse binary rewards and follow a Multi-Goal Reinforcement Learning (RL) framework in which an agent is told what to do using an additional input. The second part of the paper presents a set of concrete research ideas for improving RL algorithms, most of which are related to Multi-Goal RL and Hindsight Experience Replay.
Compounds of unknown constitution, analyzing as anhydrides, have often been reported in dealing with anthranilic acids, salicylic acids, and related compounds1. While working with N-phenylanthranilic acid and N, N-diphenylanthranilic acid the authors observed that the action of benzenesulfonyl chloride on pyridine solutions of these acids did not lead to N-benzenesulfonyl derivatives of the acids, but to yellow anhydrides. These anhydrides were observed to give rise to acridones on heating, and to rearrange to compounds of a peptide type if a hydrogen atom was still available on the nitrogen for rearrangement. To the yellow anhydride, by virtue of its reaction with ammonia and alkalies, was acribed the acyl anhydride formula2, hitherto unrecognized among such amino-acids. The present communication describes further results with substituted IV-arylanthranilic acids. The reaction appears to be a general one for N-arylanthranilic acids. Attempts have been made to obtain the peptide type of compound from JV-arylanthranilic acids by other methods not involving the possibility of anhydride formation in the synthesis. These have been so far unsuccessful and the preparation of JV-arylanthranoyl-JV-arylanthranilic acids is at present limited to the method described above.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.