3 papers
cs.AR2025
BitParticle: Partializing Sparse Dual-Factors to Build Quasi-Synchronizing MAC Arrays for Energy-efficient DNNs
Feilong Qiaoyuan, Jihe Wang, Zhiyu Sun +3
Bit-level sparsity in quantized deep neural networks (DNNs) offers significant potential for optimizing Multiply-Accumulate (MAC) operations. However, two key challenges still limi…
cs.DC2024
A Transverse-Read-assisted Fast Valid-Bits Collection in Stochastic Computing MACs for Energy-Efficient in-RTM DNNs
Jihe Wang, Zhiying Zhang, Xingwu Dong +1
It looks very attractive to coordinate racetrack-memory (RM) and stochastic-computing (SC) jointly to build an ultra-low power neuron-architecture.However, the above combination ha…
cs.LG2024
CiliaGraph: Enabling Expression-enhanced Hyper-Dimensional Computation in Ultra-Lightweight and One-Shot Graph Classification on Edge
Yuxi Han, Jihe Wang, Danghui Wang
Graph Neural Networks (GNNs) are computationally demanding and inefficient when applied to graph classification tasks in resource-constrained edge scenarios due to their inherent p…