More Like this

1. Limits of Data Driven Steganography Detectors

   https://doi.org/10.1145/3577163.3595094  

   Kaziakhmedov, E.; Dworetzky, E.; Fridrich, J. (January 2023, ACM Information Hiding and Multimedia Security Workshop)

   While deep learning has revolutionized image steganalysis in terms of performance, little is known about how much modern data-driven detectors can still be improved. In this paper, we approach this difficult and currently wide open question by working with artificial but realistic looking images with a known statistical model that allows us to compute the detectability of modern content-adaptive algorithms with respect to the most powerful detectors. Multiple artificial image datasets are crafted with different levels of content complexity and noise power to assess their influence on the gap between both types of detectors. Experiments with SRNet as the heuristic detector indicate that in dependent noise contributes less to the performance gap than content of the same MSE. While this loss is rather small for smooth images, it can be quite large for textured images. A network trained on many realizations of a fixed textured scene will, however, recuperate most of the loss, suggesting that networks have the capacity to approximately learn the parameters of a cover source narrowed to a fixed scene. 

   more » « less
2. Limits of Data Driven Steganography Detectors

   Dworetzky, E. (June 2023, ACM Information Hiding and Multimedia Security Workshop)

   While deep learning has revolutionized image steganalysis in terms of performance, little is known about how much modern data-driven detectors can still be improved. In this paper, we approach this difficult and currently wide open question by working with artificial but realistic looking images with a known statistical model that allows us to compute the detectability of modern content-adaptive algorithms with respect to the most powerful detectors. Multiple artificial image datasets are crafted with different levels of content complexity and noise power to assess their influence on the gap between both types of detectors. Experiments with SRNet as the heuristic detector indicate that independent noise contributes less to the performance gap than content of the same MSE. While this loss is rather small for smooth images, it can be quite large for textured images. A network trained on many realizations of a fixed textured scene will, however, recuperate most of the loss, suggesting that networks have the capacity to approximately learn the parameters of a cover source narrowed to a fixed scene. 

   more » « less
3. TapWeight: Reweighting Pretraining Objectives for Task-Adaptive Pretraining

   Zhang, Ruiyi; Somayajula, Sai Ashish; Xie, Pengtao (June 2025, Transactions on machine learning research)

   Large-scale general domain pretraining followed by downstream-specific finetuning has become a predominant paradigm in machine learning. However, discrepancies between the pretraining and target domains can still lead to performance degradation in certain cases, underscoring the need for task-adaptive continued pretraining (TAP). TAP methods typically involve continued pretraining on task-specific unlabeled datasets or introducing additional unsupervised learning objectives to enhance model capabilities. While many TAP methods perform continued pretraining with multiple pretraining objectives, they often determine the tradeoff parameters between objectives manually, resulting in suboptimal outcomes and higher computational costs. In this paper, we propose TapWeight, a task-adaptive pretraining framework which automatically determines the optimal importance of each pretraining objective based on downstream feedback. TapWeight reweights each pretraining objective by solving a multi-level optimization problem. We applied TapWeight to both molecular property prediction and natural language processing tasks, significantly surpassing baseline methods. Experimental results validate the effectiveness and generalizability of TapWeight. 

   more » « less
4. ImageNet Pre-trained CNNs for JPEG Steganalysis

   https://doi.org/10.1109/WIFS49906.2020.9360897  

   Yousfi, Yassine; Butora, Jan; Khvedchenya, Eugene; Fridrich, Jessica (December 2020, 12th IEEE Workshop on Information Security and Forensics)

   null (Ed.)

   In this paper, we study the EfficientNet family pre-trained on ImageNet when used for steganalysis using transfer learning. We show that certain “surgical modifications” aimed at maintaining the input resolution in EfficientNet architectures significantly boost their performance in JPEG steganalysis, establishing thus new benchmarks. The modified models are evaluated by their detection accuracy, the number of parameters, the memory consumption, and the total floating point operations (FLOPs) on the ALASKA II dataset. We also show that, surprisingly, EfficientNets in their “vanilla form” do not perform as well as the SRNet in BOSSbase+BOWS2. This is because, unlike ALASKA II images, BOSSbase+BOWS2 contains aggressively subsampled images with more complex content. The surgical modifications in EfficientNet remedy this underperformance as well. 

   more » « less
5. On Comparing Ad Hoc Detectors with Statistical Hypothesis Tests

   Dworetzky, E. (June 2023, ACM Information Hiding and Multimedia Security WOrkshop)

   This paper addresses how to fairly compare ROCs of ad hoc (or data driven) detectors with tests derived from statistical models of digital media. We argue that the ways ROCs are typically drawn for each detector type correspond to different hypothesis testing problems with different optimality criteria, making the ROCs incomparable. To understand the problem and why it occurs, we model a source of natural images as a mixture of scene oracles and derive optimal detectors for the task of image steganalysis. Our goal is to guarantee that, when the data follows the statistical model adopted for the hypothesis test, the ROC of the optimal detector bounds the ROC of the ad hoc detector. While the results are applicable beyond the field of image steganalysis, we use this setup to point out possi- ble inconsistencies when comparing both types of detectors and explain guidelines for their proper comparison. Experiments on an artificial cover source with a known model with real stegano- graphic algorithms and deep learning detectors are used to confirm our claims. 

   more » « less