More Like this

1. Leave It to Weaver

   https://doi.org/10.1145/3289239.3291459  

   Mahoney, William; Franco, Joseph; Hoff, Greg; McDonald, J. Todd (December 2018, Proceedings of the 8th Software Security, Protection, and Reverse Engineering Workshop (SSPREW))

   Malware authors make use of several techniques to obfuscate code from reverse engineering tools such as IdaPro. Typically, these techniques tend to be effective for about three to six instructions, but eventually the tools can properly disassemble the remaining code once the tool is again synchronized with the operation codes. But this loss of synchronization can be used to hide information within the instructions – steganography. Our research explores an approach to this by presenting “Weaver”, a framework for executable steganography. “Weaver” differs from other techniques in how it hides malicious instructions: the hiding instructions are prepared by generating an assembly listing of the program and finding candidate hiding locations, the steganography instructions are prepared by creating an assembly listing of the program to obtain the operation codes to be hidden, and the “weaving” process merges the two. This “weaving” attempts to place all the steganography instructions into candidate locations found in the hiding instructions. 

   more » « less
2. Regularization-wise double descent: Why it occurs and how to eliminate it

   Fatih Furkan Yilmaz and Reinhard Heckel (September 2022, IEEE International Symposium on Information Theory)

   The risk of overparameterized models, in particular deep neural networks, is often double- descent shaped as a function of the model size. Recently, it was shown that the risk as a function of the early-stopping time can also be double-descent shaped, and this behavior can be explained as a super-position of bias-variance tradeoffs. In this paper, we show that the risk of explicit L2-regularized models can exhibit double descent behavior as a function of the regularization strength, both in theory and practice. We find that for linear regression, a double descent shaped risk is caused by a superposition of bias-variance tradeoffs corresponding to different parts of the model and can be mitigated by scaling the regularization strength of each part appropriately. Motivated by this result, we study a two-layer neural network and show that double descent can be eliminated by adjusting the regularization strengths for the first and second layer. Lastly, we study a 5-layer CNN and ResNet-18 trained on CIFAR-10 with label noise, and CIFAR-100 without label noise, and demonstrate that all exhibit double descent behavior as a function of the regularization strength. 

   more » « less
3. SURPRISE! and When to Schedule It.

   https://doi.org/10.24963/ijcai.2021/36  

   Huang, Zhihuan; Xu, Shengwei; Shan, You; Lu, Yuxuan; Kong, Yuqing; Liu, Tracy Xiao; Schoenebeck, Grant (August 2021, Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence)

   Information flow measures, over the duration of a game, the audience’s belief of who will win, and thus can reflect the amount of surprise in a game. To quantify the relationship between information flow and audiences' perceived quality, we conduct a case study where subjects watch one of the world’s biggest esports events, LOL S10. In addition to eliciting information flow, we also ask subjects to report their rating for each game. We find that the amount of surprise in the end of the game plays a dominant role in predicting the rating. This suggests the importance of incorporating when the surprise occurs, in addition to the amount of surprise, in perceived quality models. For content providers, it implies that everything else being equal, it is better for twists to be more likely to happen toward the end of a show rather than uniformly throughout. 

   more » « less
4. “Dump it, Destroy it, Send it to Data Heaven”: Blind People’s Expectations for Visual Privacy in Visual Assistance Technologies

   https://doi.org/10.1145/3587281.3587296  

   Stangl, Abigale; Sadjo, Emma; Emami-Naeini, Pardis; Wang, Yang; Gurari, Danna; Findlater, Leah (April 2023, “Dump it, Destroy it, Send it to Data Heaven”: Blind People’s Expectations for Visual Privacy in Visual Assistance Technologies)
5. Build It, Break It, Fix It: Contesting Secure Development

   https://doi.org/10.1145/3383773  

   Parker, James; Hicks, Michael; Ruef, Andrew; Mazurek, Michelle L.; Levin, Dave; Votipka, Daniel; Mardziel, Piotr; Fulton, Kelsey R. (May 2020, ACM Transactions on Privacy and Security)