Guillaume Garreau

About

Guillaume Garreau is a leading researcher in the design of efficient, hardware-oriented machine learning systems, with a primary focus on the FPGA implementation of deep neural networks. His most-cited work, a 2015 study on implementing a Deep Belief Network (DBN) for character recognition using stochastic computation, has garnered 45 citations. This seminal contribution demonstrates his expertise in bridging the gap between complex deep learning architectures and resource-constrained hardware, pioneering the use of stochastic computing to drastically reduce the power and area required for neural network inference. By enabling sophisticated models like DBNs to run on reconfigurable logic, Garreau’s research directly addresses the critical challenge of deploying AI in embedded and edge devices. His work is highly influential in the fields of hardware acceleration, neuromorphic computing, and low-power AI, providing a practical pathway for integrating advanced classification and generative tasks—from vision to robotics—into real-world, energy-efficient systems.

Research Focus

Key Achievements

Top Papers

1. 1
   FPGA implementation of a Deep Belief Network architecture for character recognition using stochastic computation

   45 citations · 2015

Key Collaborators

Contact & Links