Picture for Maksim Jenihhin

Maksim Jenihhin

FORTALESA: Fault-Tolerant Reconfigurable Systolic Array for DNN Inference

Add code
Mar 06, 2025
Viaarxiv icon

DeepVigor+: Scalable and Accurate Semi-Analytical Fault Resilience Analysis for Deep Neural Network

Add code
Oct 21, 2024
Viaarxiv icon

ProAct: Progressive Training for Hybrid Clipped Activation Function to Enhance Resilience of DNNs

Add code
Jun 10, 2024
Viaarxiv icon

Cost-Effective Fault Tolerance for CNNs Using Parameter Vulnerability Based Hardening and Pruning

Add code
May 17, 2024
Viaarxiv icon

AdAM: Adaptive Fault-Tolerant Approximate Multiplier for Edge DNN Accelerators

Add code
Mar 05, 2024
Viaarxiv icon

SAFFIRA: a Framework for Assessing the Reliability of Systolic-Array-Based DNN Accelerators

Add code
Mar 05, 2024
Figure 1 for SAFFIRA: a Framework for Assessing the Reliability of Systolic-Array-Based DNN Accelerators
Figure 2 for SAFFIRA: a Framework for Assessing the Reliability of Systolic-Array-Based DNN Accelerators
Figure 3 for SAFFIRA: a Framework for Assessing the Reliability of Systolic-Array-Based DNN Accelerators
Figure 4 for SAFFIRA: a Framework for Assessing the Reliability of Systolic-Array-Based DNN Accelerators
Viaarxiv icon

Exploration of Activation Fault Reliability in Quantized Systolic Array-Based DNN Accelerators

Add code
Jan 17, 2024
Viaarxiv icon

Enhancing Fault Resilience of QNNs by Selective Neuron Splitting

Add code
Jun 16, 2023
Viaarxiv icon

APPRAISER: DNN Fault Resilience Analysis Employing Approximation Errors

Add code
May 31, 2023
Viaarxiv icon

A Novel Fault-Tolerant Logic Style with Self-Checking Capability

Add code
May 31, 2023
Viaarxiv icon