Tensordyne Revives Logarithmic Math In A Bid To Cut AI Power Use

Tensordyne says logarithmic computing could reduce AI inference costs and power demands, offering an alternative to conventional chip designs.
IEEE

HyMM: A Hybrid Sparse-Dense Matrix Multiplication Accelerator for GCNs

Abstract: Graph convolutional networks (GCNs) are emerging neural network models designed to process graph-structured data. Due to massively parallel computations using irregular data structures by ...
IEEE

Phase Shift Matrix Design With Vector Quantization in RIS-Aided Physical Layer Key Generation

Abstract: The key generation rate (KGR) of physical layer key generation (PLKG) is severely restricted in quasi-static environments due to a lack of entropy. To address the problem, a reconfigurable ...