Google Neural Network Models for Edge Devices: Analyzing & Mitigating ML Inference Bottlenecks; PACT

Talk Title: Google Neural Network Models for Edge Devices: Analyzing and Mitigating Machine Learning Inference Bottlenecks International Conference on Parallel Architectures and Compilation Techniques (PACT), Virtual, September 2021, Session 3: Characterization and Near-Memory Computing. Speaker: Geraldo F. Oliveira, SAFARI Research Group, ETH Zurich Duration: 14 minutes Full paper (PDF): Slides (PPTX): Slides (PDF): Abstract: Emerging edge computing platforms often contain machine learning (ML) accelerators that can accelerate inference for a wide range of neural network (NN) models. These models are designed to fit within the limited area and energy c