no code implementations • 25 Jan 2023 • Ayesha Siddique, Ripan Kumar Kundu, Gautam Raj Mode, Khaza Anuarul Hoque
We observe that approximate adversarial training can significantly improve the robustness of PdM models (up to 54X) and outperforms the state-of-the-art PdM defense methods by offering 3X more robustness.
no code implementations • 12 Jan 2023 • Syed Tihaam Ahmad, Ayesha Siddique, Khaza Anuarul Hoque
Therefore, researchers in the recent past have extensively studied the robustness and defense of DNNs and SNNs under adversarial attacks.
no code implementations • 12 Jan 2023 • Ayesha Siddique, Khaza Anuarul Hoque
Our proposed FalVolt mitigation method improves the performance of systolicSNNs by enabling them to operate at fault rates of up to 60\%, with a negligible drop in classification accuracy (as low as 0. 1\%).
no code implementations • 2 Dec 2021 • Ayesha Siddique, Khaza Anuarul Hoque
Approximate computing is known for its effectiveness in improvising the energy efficiency of deep neural network (DNN) accelerators at the cost of slight accuracy loss.
no code implementations • 8 Jan 2021 • Ayesha Siddique, Kanad Basu, Khaza Anuarul Hoque
Our quantitative analysis shows that the permanent faults exacerbate the accuracy loss in AxDNNs when compared to the accurate DNN accelerators.