VECODI introduces SHANGRI-LA, an intermediate-privilege runtime on TrustZone-M, to enable verifiable confidential DNN inference on constrained edge devices with small TCB and overhead.
An evaluation of edge tpu accelerators for con- volutional neural networks,
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
Privatar uses horizontal frequency partitioning and distribution-aware minimal perturbation to enable private offloading of VR avatar reconstruction, supporting 2.37x more users with modest overhead.
Demonstrates single-shot 3D matrix-matrix multiplication in an optical tensor processor that accelerates CNNs and DNNs at 20 aJ per MAC with 96.4% accuracy on image recognition using 292616 parameters.
citing papers explorer
-
Single-Shot Matrix-Matrix Multiplication Optical Tensor Processor for Deep Learning
Demonstrates single-shot 3D matrix-matrix multiplication in an optical tensor processor that accelerates CNNs and DNNs at 20 aJ per MAC with 96.4% accuracy on image recognition using 292616 parameters.