Search Results for author:
Found 5 papers, 2 papers with code
MaiT: integrating spatial locality into image transformers with attention masks
Moreover, attention masks regulate the training of attention maps, which facilitates the convergence and improves the accuracy of deeper transformers.
FORMS: Fine-grained Polarized ReRAM-based In-situ Computation for Mixed-signal DNN Accelerator
With weights stored in the ReRAM crossbar cells as conductance, when the input vector is applied to word lines, the matrix-vector multiplication results can be generated as the current in bit lines.
Rethinking Floating Point Overheads for Mixed Precision DNN Accelerators
We present novel optimizations based on the above observations to reduce the FP arithmetic hardware overheads.
Post-Training Piecewise Linear Quantization for Deep Neural Networks
Quantization plays an important role in the energy-efficient deployment of deep neural networks on resource-limited devices.
Newton: Gravitating Towards the Physical Limits of Crossbar Acceleration
Many recent works have designed accelerators for Convolutional Neural Networks (CNNs).
