Manuel Eberhardinger
Attending the 17th International Workshop on
Neural-Symbolic Learning and Reasoning
The 17th International Workshop on Neural-Symbolic Learning and Reasoning took place this year in Italy. Our PhD student Manuel Eberhardinger presented the paper “Towards Explainable Decision Making with Neural Program Synthesis and Library Learning.
The workshop was held near Siena in an old monastery, where the latest research on neurosymbolic artificial intelligence was presented and discussed. There were also keynote presentations by two Turing award laureates. Yann LeCun, who received the 2018 Turing Award for significant breakthroughs in Deep Learning, presented his vision for how to develop more human-like AI systems. Leslie Valiant received the Turing Award in 2010 for transformative contributions to the theory of computation, including the theory of probably approximately correct (PAC) learning. His keynote addressed how learning can be augmented by reasoning.
Abstract
Generating explanations for the behavior of artificial or living agents is still an underexplored problem. We propose a new framework to make agent behavior explainable based on program synthesis. Programs have the advantage that they are inherently interpretable and can be verified for correctness. In detail, we use neural program synthesis with a language model in combination with library learning. In addition, we introduce a domain-agnostic curriculum that is necessary for the system to learn at all. We compare our methods with traditional and state-of-the-art program synthesis systems and justify the necessity of the curriculum and library learning module in ablation studies, followed by an analysis of the extracted libraries.
Authors: Manuel Eberhardinger, Johannes Maucher, Setareh Maghsudi