Project Details

LEDNeCo: Low Energy Deep Neurocomputing

Project Period: 1. 1. 2025 - 31. 12. 2027

Project Type: grant

Code: GA25-15490S

Agency: Czech Science Foundation

Program: Standardní projekty

English title
LEDNeCo: Low Energy Deep Neurocomputing
Type
grant
Keywords

deep neural networks;tranformers;energy complexity;hardware accelerator;approximation theory;robust learning;genetic programming

Abstract

Modern artificial intelligence technologies based on deep neural networks (DNNs) such as GPT are computationally extremely demanding. In addition to consuming an enormous amount of energy, this limits their deployment in battery-powered embedded (edge) devices (e.g. smart mobile apps). LEDNeCo is a project of basic research whose ambition is to develop a low-energy deep neurocomputing paradigm based on machine-independent energy complexity theory for DNNs, which issues from practical experience in the design of diverse DNN hardware accelerators. Among other things, universal lower bounds on energy complexity of DNNs and estimates of inference error will be derived for identifying DNN components (e.g. weights, neurons, layers) whose approximation is provably the most energy efficient. The achieved theoretical knowledge will be used in new advanced approximation techniques (e.g. weight compression, Boolean optimization, robust aproximation of components) for low-power hardware implementations of DNN (incl. transformer), which will be tested on benchmark datasets.

Team members
Vašíček Zdeněk, doc. Ing., Ph.D. (DCSY FIT BUT) , research leader
Klhůfek Jan, Ing. (DCSY FIT BUT)
Mrázek Vojtěch, Ing., Ph.D. (DCSY FIT BUT)
Sekanina Lukáš, prof. Ing., Ph.D. (DCSY FIT BUT)
Publications

2025

  • MRÁZEK Vojtěch, BALASKAS Konstantinos, DUARTE Carolina Lozano Paula, VAŠÍČEK Zdeněk, TAHOORI Mehdi and ZERVAKIS Georgios. Arbitrary Precision Printed Ternary Neural Networks with Holistic Evolutionary Approximation. IEEE Transactions on Circuits and Systems for Artificial Intelligence, 2025, pp. 1-13. ISSN 2996-6647. Detail
  • KALKREUTH Roman, DE França Fabricio Olivetti, JANKOVIC Anja, ANASTACIO Marie, DIERKES Julian, VAŠÍČEK Zdeněk and HOOS Holger. TinyverseGP: Towards a Modular Cross-domain Benchmarking Framework for Genetic Programming. In: Proceedings of the Genetic and Evolutionary Computation Conference Companion. Malaga: Association for Computing Machinery, 2025, pp. 2172-2176. ISBN 979-8-4007-1464-1. Detail
  • PLEVAČ Lukáš and VAŠÍČEK Zdeněk. Towards Efficient Semantic Mutation in CGP: Enhancing SOMOk. In: Proceedings of the Genetic and Evolutionary Computation Conference Companion. Malaga: Association for Computing Machinery, 2025, pp. 2172-2176. ISBN 979-8-4007-1464-1. Detail
  • SEDLÁK David, KLHŮFEK Jan, MRÁZEK Vojtěch and VAŠÍČEK Zdeněk. Towards Efficient Scheduling of Transformer Neural Network Computation for Edge AI Deployment. In: Proceedings of the Genetic and Evolutionary Computation Conference Companion. Malaga: Association for Computing Machinery, 2025, pp. 2242-2248. ISBN 979-8-4007-1464-1. Detail
Back to top