Changes between Version 20 and Version 21 of Other/Summer/2020/AdvML
- Timestamp:
- Jul 6, 2020, 4:17:36 AM (4 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Other/Summer/2020/AdvML
v20 v21 28 28 - Statistical Pooling: https://www.tensorflow.org/api_docs/python/tf/nn/moments 29 29 - Probabilistic Linear Discriminant Analysis for Inferences About Identity: https://www.orbit-lab.org/attachment/wiki/Other/Summer/2020/AdvML/Probabilistic%20Linear%20Discriminant%20Analysis%20for%20Inferences%20About%20Identity.pdf 30 *Week 6 31 - Introduction of Fast Gradient Sign Method (FSGM): https://towardsdatascience.com/perhaps-the-simplest-introduction-of-adversarial-examples-ever-c0839a759b8d#:~:text=Fast%20Gradient%20Sign%20Method%20(FGSM)&text=In%20essence%2C%20FGSM%20is%20to,small%20number%20via%20max%20norm. 32 - Adversarial example using FGSM: https://www.tensorflow.org/tutorials/generative/adversarial_fgsm 33 - Cross-entropy cost function: https://eng.libretexts.org/Bookshelves/Computer_Science/Book%3A_Neural_Networks_and_Deep_Learning_(Nielsen)/03%3A_Improving_the_way_neural_networks_learn/3.01%3A_The_cross-entropy_cost_function 30 34 31 35 == Reading Material == … … 73 77 - Read the paper: Practical Adversarial Attacks Against Speaker Recognition Systems (HotMobile’20) and get familiar with the untargeted attack 74 78 79 == Week 6 Activities == 80 - Develop an untargeted attack that can generate adversarial samples based on the sample code and tutorial. \\ 81 -- Understand Fast Gradient Sign Method (FSGM) \\ 82 -- Understand cross-entropy as cost function \\ 83 - Evaluate the performance of the adversarial samples on the voice assistant system (X-Vector). 84 75 85 == Project Website == 76 86 - [https://chunnubansal.wixsite.com/winlab-amlavas]
