IEEE

Panacea: Novel DNN Accelerator using Accuracy-Preserving Asymmetric Quantization and Energy-Saving Bit-Slice Sparsity

Abstract: Low bit-precisions and their bit-slice sparsity have recently been studied to accelerate general matrix-multiplications (GEMM) during large-scale deep neural network (DNN) inferences. While ...
IEEE

Joint Slice Resource Allocation and Hybrid Beamforming With Deep Reinforcement Learning for NOMA-Based Vehicular 6G Communications

Abstract: The escalating demand for high data rates and dependable communications in forthcoming wireless networks has led to the exploration of innovative solutions. Among these, the fusion of ...