Gegelati: Lightweight Artificial Intelligence through Generic and Evolvable Tangled Program Graphs
Karol Desnos (UNIV-RENNES, INSA Rennes, IETR), Nicolas Sourbier (INSA, Rennes, UNIV-RENNES, IETR), Pierre-Yves Raumer (INSA Rennes, IETR), Olivier, Gesny, Maxime Pelcat (UNIV-RENNES, INSA Rennes, IETR)

TL;DR
Gegelati introduces a lightweight reinforcement learning library based on Tangled Program Graphs, optimized for embedded systems, with parallel training and customizable instruction sets demonstrated on various hardware architectures.
Contribution
The paper presents Gegelati, a library enabling scalable, parallel training of TPGs with customizable instructions, suitable for resource-constrained embedded systems.
Findings
Parallel training scales from high-end to low-power MPSoCs.
Customizable instructions influence training outcomes.
TPGs achieve comparable performance to DNNs with less resource use.
Abstract
Tangled Program Graph (TPG) is a reinforcement learning technique based on genetic programming concepts. On state-of-the-art learning environments, TPGs have been shown to offer comparable competence with Deep Neural Networks (DNNs), for a fraction of their computational and storage cost. This lightness of TPGs, both for training and inference, makes them an interesting model to implement Artificial Intelligences (AIs) on embedded systems with limited computational and storage resources. In this paper, we introduce the Gegelati library for TPGs. Besides introducing the general concepts and features of the library, two main contributions are detailed in the paper: 1/ The parallelization of the deterministic training process of TPGs, for supporting heterogeneous Multiprocessor Systems-on-Chips (MPSoCs). 2/ The support for customizable instruction sets and data types within the genetically…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
