no code implementations • 19 Apr 2024 • Zhixin Pan, Emma Andrews, Laura Chang, Prabhat Mishra
Data augmentation is widely used to mitigate data bias in the training dataset.
no code implementations • 17 Apr 2024 • Ziyu Shu, Zhixin Pan
SDIP efficiently utilizes this trait in a reinforcement learning manner, where the current iteration's output is utilized by a steering algorithm to update the network input for the next iteration, guiding the algorithm toward improved results.
no code implementations • 4 May 2023 • Zhixin Pan, Prabhat Mishra
Extensive experimental evaluation demonstrates that proposed approach deployed on TPU can provide drastic improvement in interpretation time (39x on average) as well as energy efficiency (69x on average) compared to existing acceleration techniques.
Explainable artificial intelligence Explainable Artificial Intelligence (XAI)
no code implementations • 18 May 2022 • Zhixin Pan, Prabhat Mishra
In this paper, we propose a novel backdoor attack based on effective learning and targeted utilization of reverse distribution.
no code implementations • 22 Mar 2021 • Zhixin Pan, Prabhat Mishra
(1) To the best of our knowledge, our proposed work is the first attempt in enabling hardware acceleration of explainable ML using TPUs.
no code implementations • 22 Mar 2021 • Zhixin Pan, Prabhat Mishra
One promising strategy to counter adversarial attacks is to utilize spectral normalization, which ensures that the trained model has low sensitivity towards the disturbance of input samples.