Papers and Publications

This is an abridged list of all of Yang Katie Zhao's publications. A complete list of publications and citations are available on Google Scholar.

Expand all

2023

Instant-NeRF: Instant On-Device Neural Radiance Field Training via Algorithm-Accelerator Co-Designed Near-Memory Processing

Instant-3d: Instant neural radiance field training towards on-device ar/vr 3d reconstruction

Vitcod: Vision transformer acceleration via dedicated algorithm and accelerator co-design

2022

RT-NeRF: Real-time on-device neural radiance fields towards immersive AR/VR rendering

NASA: Neural architecture search and acceleration for hardware inspired hybrid networks

EyeCoD: eye tracking system acceleration via flatcam-based algorithm & accelerator co-design

e-G2C: A 0.14-to-8.31 µJ/Inference NN-based Processor with Continuous On-chip Adaptation for Anomaly Detection and ECG Conversion from EGM

I-FlatCam: A 253 FPS, 91.49 µJ/frame ultra-compact intelligent lensless camera for real-time and efficient eye tracking in VR/AR

SmartDeal: Remodeling Deep Network Weights for Efficient Inference and Training

2021

2-in-1 accelerator: Enabling random precision switch for winning both adversarial robustness and efficiency

Hw-nas-bench: Hardware-aware neural architecture search benchmark

2020

A new MRAM-based process in-memory accelerator for efficient neural network training with floating point precision

PRISM: Energy-Efficient Polymorphic Operation Based on Spin-Orbit Torque Memory for Reconfigurable Computing

Timely: Pushing data movements and interfaces in pim accelerators towards local and in time domain

Smartexchange: Trading higher-cost memory storage/access for lower-cost computation

Practical attacks on deep neural networks by memory trojaning

Dnn-chip predictor: An analytical performance predictor for dnn accelerators with various dataflows and hardware architectures

Autodnnchip: An automated dnn chip predictor and builder for both fpgas and asics

Fractrain: Fractionally squeezing bit savings both temporally and spatially for efficient dnn training

E2-Train: Training State-of-the-art CNNs with Over 80% Less Energy