Introduction - Session Recordings and References Approximate Computing Multiprocessing, multithreading and vectorization GPU architecture and optimizations Floating-point, fixed-point and integer formats Roofline model and arithmetic intensity Common optimizations Cache Blocking Four convolution strategies: Direct, GEMM, FFT and Winograd Model-size aware and system-aware pruning of CNNs Deep Learning on Systolic Array and TPU Deep Learning on FPGA Deep Learning on CPU Deep Learning on Xeon Phi Deep Learning on Mobile GPU Deep Learning on Smartphone Deep Learning on Cerebras Wafer Scale Engine Deep Learning on Memristors Comparative evaluation of CPU/GPU/FPGA/ASIC for accelerating autonomous driving MLPerf Benchmark for Evaluating DNN Accelerators Security of DNN Algorithms and Accelerators Reliability of DNN Algorithms and Accelerators DNN Training on a single GPU Distributed Training of DNN on multiple GPU nodes Neural Branch Predictor and Its Application in Brain-Implant Deep Learning on Graphcore Intelligence Processing Unit (IPU) Energy management approaches Deep Learning on Intel Habana Goya and Gaudi Processors Deep Learning on Qualcomm’s Hexagon DSP Deep Learning on Alibaba HanGuang Processor Deep Learning on Untether’s TsunAImi Processor Deep Learning on Groq’s Tensor Streaming Processor (TSP) Deep Learning on TensTorrent’s Grayskull Processor