Picture for Xiaobo Sharon Hu

Xiaobo Sharon Hu

A Remedy to Compute-in-Memory with Dynamic Random Access Memory: 1FeFET-1C Technology for Neuro-Symbolic AI

Add code
Oct 20, 2024
Viaarxiv icon

Compute-in-Memory based Neural Network Accelerators for Safety-Critical Systems: Worst-Case Scenarios and Protections

Add code
Dec 11, 2023
Viaarxiv icon

Improving Realistic Worst-Case Performance of NVCiM DNN Accelerators through Training with Right-Censored Gaussian Noise

Add code
Jul 29, 2023
Figure 1 for Improving Realistic Worst-Case Performance of NVCiM DNN Accelerators through Training with Right-Censored Gaussian Noise
Figure 2 for Improving Realistic Worst-Case Performance of NVCiM DNN Accelerators through Training with Right-Censored Gaussian Noise
Figure 3 for Improving Realistic Worst-Case Performance of NVCiM DNN Accelerators through Training with Right-Censored Gaussian Noise
Figure 4 for Improving Realistic Worst-Case Performance of NVCiM DNN Accelerators through Training with Right-Censored Gaussian Noise
Viaarxiv icon

On the Viability of using LLMs for SW/HW Co-Design: An Example in Designing CiM DNN Accelerators

Add code
Jun 12, 2023
Viaarxiv icon

Negative Feedback Training: A Novel Concept to Improve Robustness of NVCiM DNN Accelerators

Add code
May 23, 2023
Viaarxiv icon

ApproxTrain: Fast Simulation of Approximate Multipliers for DNN Training and Inference

Add code
Sep 13, 2022
Figure 1 for ApproxTrain: Fast Simulation of Approximate Multipliers for DNN Training and Inference
Figure 2 for ApproxTrain: Fast Simulation of Approximate Multipliers for DNN Training and Inference
Figure 3 for ApproxTrain: Fast Simulation of Approximate Multipliers for DNN Training and Inference
Figure 4 for ApproxTrain: Fast Simulation of Approximate Multipliers for DNN Training and Inference
Viaarxiv icon

Computing-In-Memory Neural Network Accelerators for Safety-Critical Systems: Can Small Device Variations Be Disastrous?

Add code
Jul 15, 2022
Figure 1 for Computing-In-Memory Neural Network Accelerators for Safety-Critical Systems: Can Small Device Variations Be Disastrous?
Figure 2 for Computing-In-Memory Neural Network Accelerators for Safety-Critical Systems: Can Small Device Variations Be Disastrous?
Figure 3 for Computing-In-Memory Neural Network Accelerators for Safety-Critical Systems: Can Small Device Variations Be Disastrous?
Figure 4 for Computing-In-Memory Neural Network Accelerators for Safety-Critical Systems: Can Small Device Variations Be Disastrous?
Viaarxiv icon

SWIM: Selective Write-Verify for Computing-in-Memory Neural Accelerators

Add code
Feb 17, 2022
Figure 1 for SWIM: Selective Write-Verify for Computing-in-Memory Neural Accelerators
Figure 2 for SWIM: Selective Write-Verify for Computing-in-Memory Neural Accelerators
Figure 3 for SWIM: Selective Write-Verify for Computing-in-Memory Neural Accelerators
Viaarxiv icon

Deep Random Forest with Ferroelectric Analog Content Addressable Memory

Add code
Oct 06, 2021
Figure 1 for Deep Random Forest with Ferroelectric Analog Content Addressable Memory
Figure 2 for Deep Random Forest with Ferroelectric Analog Content Addressable Memory
Figure 3 for Deep Random Forest with Ferroelectric Analog Content Addressable Memory
Figure 4 for Deep Random Forest with Ferroelectric Analog Content Addressable Memory
Viaarxiv icon

RADARS: Memory Efficient Reinforcement Learning Aided Differentiable Neural Architecture Search

Add code
Sep 13, 2021
Figure 1 for RADARS: Memory Efficient Reinforcement Learning Aided Differentiable Neural Architecture Search
Figure 2 for RADARS: Memory Efficient Reinforcement Learning Aided Differentiable Neural Architecture Search
Figure 3 for RADARS: Memory Efficient Reinforcement Learning Aided Differentiable Neural Architecture Search
Figure 4 for RADARS: Memory Efficient Reinforcement Learning Aided Differentiable Neural Architecture Search
Viaarxiv icon