Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310746648> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4310746648 abstract "Deep neural networks have emerged as the workhorse for a large section of robotics and control applications, especially as models for dynamical systems. Such data-driven models are in turn used for designing and verifying autonomous systems. They are particularly useful in modeling medical systems where data can be leveraged to individualize treatment. In safety-critical applications, it is important that the data-driven model is conformant to established knowledge from the natural sciences. Such knowledge is often available or can often be distilled into a (possibly black-box) model. For instance, an F1 racing car should conform to Newton's laws (which are encoded within a unicycle model). In this light, we consider the following problem - given a model $M$ and a state transition dataset, we wish to best approximate the system model while being a bounded distance away from $M$. We propose a method to guarantee this conformance. Our first step is to distill the dataset into a few representative samples called memories, using the idea of a growing neural gas. Next, using these memories we partition the state space into disjoint subsets and compute bounds that should be respected by the neural network in each subset. This serves as a symbolic wrapper for guaranteed conformance. We argue theoretically that this only leads to a bounded increase in approximation error; which can be controlled by increasing the number of memories. We experimentally show that on three case studies (Car Model, Drones, and Artificial Pancreas), our constrained neurosymbolic models conform to specified models (each encoding various constraints) with order-of-magnitude improvements compared to the augmented Lagrangian and vanilla training methods. Our code can be found at: https://github.com/kaustubhsridhar/Constrained_Models" @default.
- W4310746648 created "2022-12-17" @default.
- W4310746648 creator A5003254024 @default.
- W4310746648 creator A5030456600 @default.
- W4310746648 creator A5043161001 @default.
- W4310746648 creator A5054356243 @default.
- W4310746648 date "2022-12-02" @default.
- W4310746648 modified "2023-10-17" @default.
- W4310746648 title "Guaranteed Conformance of Neurosymbolic Models to Natural Constraints" @default.
- W4310746648 doi "https://doi.org/10.48550/arxiv.2212.01346" @default.
- W4310746648 hasPublicationYear "2022" @default.
- W4310746648 type Work @default.
- W4310746648 citedByCount "0" @default.
- W4310746648 crossrefType "posted-content" @default.
- W4310746648 hasAuthorship W4310746648A5003254024 @default.
- W4310746648 hasAuthorship W4310746648A5030456600 @default.
- W4310746648 hasAuthorship W4310746648A5043161001 @default.
- W4310746648 hasAuthorship W4310746648A5054356243 @default.
- W4310746648 hasBestOaLocation W43107466481 @default.
- W4310746648 hasConcept C105795698 @default.
- W4310746648 hasConcept C110251889 @default.
- W4310746648 hasConcept C11413529 @default.
- W4310746648 hasConcept C114614502 @default.
- W4310746648 hasConcept C119857082 @default.
- W4310746648 hasConcept C134306372 @default.
- W4310746648 hasConcept C154945302 @default.
- W4310746648 hasConcept C33923547 @default.
- W4310746648 hasConcept C34388435 @default.
- W4310746648 hasConcept C41008148 @default.
- W4310746648 hasConcept C45340560 @default.
- W4310746648 hasConcept C48103436 @default.
- W4310746648 hasConcept C50644808 @default.
- W4310746648 hasConcept C72434380 @default.
- W4310746648 hasConcept C80444323 @default.
- W4310746648 hasConceptScore W4310746648C105795698 @default.
- W4310746648 hasConceptScore W4310746648C110251889 @default.
- W4310746648 hasConceptScore W4310746648C11413529 @default.
- W4310746648 hasConceptScore W4310746648C114614502 @default.
- W4310746648 hasConceptScore W4310746648C119857082 @default.
- W4310746648 hasConceptScore W4310746648C134306372 @default.
- W4310746648 hasConceptScore W4310746648C154945302 @default.
- W4310746648 hasConceptScore W4310746648C33923547 @default.
- W4310746648 hasConceptScore W4310746648C34388435 @default.
- W4310746648 hasConceptScore W4310746648C41008148 @default.
- W4310746648 hasConceptScore W4310746648C45340560 @default.
- W4310746648 hasConceptScore W4310746648C48103436 @default.
- W4310746648 hasConceptScore W4310746648C50644808 @default.
- W4310746648 hasConceptScore W4310746648C72434380 @default.
- W4310746648 hasConceptScore W4310746648C80444323 @default.
- W4310746648 hasLocation W43107466481 @default.
- W4310746648 hasOpenAccess W4310746648 @default.
- W4310746648 hasPrimaryLocation W43107466481 @default.
- W4310746648 hasRelatedWork W1517759046 @default.
- W4310746648 hasRelatedWork W1598914363 @default.
- W4310746648 hasRelatedWork W1748450182 @default.
- W4310746648 hasRelatedWork W2275271575 @default.
- W4310746648 hasRelatedWork W2295093357 @default.
- W4310746648 hasRelatedWork W2807819528 @default.
- W4310746648 hasRelatedWork W2949390856 @default.
- W4310746648 hasRelatedWork W895230556 @default.
- W4310746648 hasRelatedWork W1580211323 @default.
- W4310746648 hasRelatedWork W1629725936 @default.
- W4310746648 isParatext "false" @default.
- W4310746648 isRetracted "false" @default.
- W4310746648 workType "article" @default.