[prof.Simon S. Woo] DASH Lab, Two Papers Approved for Publication at the CIKM 2024 International Conference

[prof.Simon S. Woo] DASH Lab, Two Papers Approved for Publication at the CIKM 2024 International Conference 1. IDENTIFY: Integral Radial and Spatial Fourier Analysis for AI-Generated Image Authentication (full paper) Writer: Inzamamul Alam, Muhammad Shahid Muneer, and Prof. Simon S. Woo This study proposes a method for detecting deepfakes generated by a new generative AI (Diffusion) technique using Integral Radial and Spatial Fourier Analysis with high performance. Notably, the proposed method shows 12-28% higher performance compared to existing approaches. IDENTIFY: Integral Radial and Spatial Fourier Analysis for AI-Generated Image Authentication, Inzamamul Alam, Muhammad Shahid Muneer, and Simon S. Woo*, 33rd ACM International Conference on Information & Knowledge Management (CIKM), Boise, Idaho, USA, October 2024 2. Deep Journey Hierarchical Attention Networks for Conversion Predictions in Digital Marketing (full paper) Writer: Girim Ban, Hyeonseok Yun, Banseok Lee, David Sung, and Prof. Simon S. Woo This study proposes Deep Journey Hierarchical Attention Networks (DJHAN) to enhance user conversion prediction in digital marketing, improving key metrics such as Conversion Rate (CVR) and Return on Ad Spend (ROAS) compared to existing methods. The proposed model demonstrated high performance when applied to real marketing data from KT/NasMedia. Deep Journey Hierarchical Attention Networks for Predictions in Digital Marketing Girim Ban, Hyeonseok Yun, Banseok Lee, David Sung, and Simon S. Woo* 33rd ACM International Conference on Information & Knowledge Management (CIKM), Boise, Idaho, USA, October 2024

Papers from Prof. Jinkyu Lee’s Lab. (RTCL@SKKU) published in ACM/IEEE DAC 2024 and IEEE RTAS 2024

A paper from RTCL@SKKU (Advisor: Jinkyu Lee) has been published in ACM/IEEE DAC 2024 and IEEE RTAS 2024. ACM/IEEE DAC 2024 Website https://www.dac.com/ IEEE RTAS 2024 Website https://2024.rtas.org/ Real-Time Computing Lab. Website https://rtclskku.github.io/website/ - Paper Title: RT-MDM: Real-Time Scheduling Framework for Multi-DNN on MCU Using External Memory - Abstract: As the application scope of DNNs executed on microcontroller units (MCUs) extends to time-critical systems, it becomes important to ensure timing guarantees for increasing demand of DNN inferences. To this end, this paper proposes RT-MDM, the first Real-Time scheduling framework for Multiple DNN tasks executed on an MCU using external memory. Identifying execution-order dependencies among segmented DNN models and memory requirements for parallel execution subject to the dependencies, we propose (i) a segment-group-based memory management policy that achieves isolated memory usage within a segment group and sharded memory usage across different segment groups, and (ii) an intra-task scheduler specialized for the proposed policy. Implementing RT-MDM on an actual system and optimizing its parameters for DNN segmentation and segment-group mapping, we demonstrate the effectiveness of RT-MDM in accommodating more DNN tasks while providing their timing guarantees. - Paper Title: RT-Swap: Addressing GPU Memory Bottlenecks for Real-Time Multi-DNN Inference - Abstract: The increasing complexity and memory demands of Deep Neural Networks (DNNs) for real-time systems pose new significant challenges, one of which is the GPU memory capacity bottleneck, where the limited physical memory inside GPUs impedes the deployment of sophisticated DNN models. This paper presents, to the best of our knowledge, the first study of addressing the GPU memory bottleneck issues, while simultaneously ensuring the timely inference of multiple DNN tasks. We propose RT-Swap, a real-time memory management framework, that enables transparent and efficient swap scheduling of memory objects, employing the relatively larger CPU memory to extend the available GPU memory capacity, without compromising timing guarantees. We have implemented RT-Swap on top of representative machine-learning frameworks, demonstrating its effectiveness in making significantly more DNN task sets schedulable at least 72% over existing approaches even when the task sets demand up to 96.2% more memory than the GPU’s physical capacity. Jinkyu Lee | jinkyu.lee@skku.edu | RTCL@SKKU | https://rtclskku.github.io/website/

DASH Lab won the Best Paper Running-Up Award (2nd Best Paper) at PAKDD 2024 in Taiwan

Binh M. Le and Simon S. Woo’s paper, “SEE: Spherical Embedding Expansion for Improving Deep Metric Learning,” received the the Best Paper Running-Up Award (2nd best paper) in PAKDD 2024 (BK CS IF=1), held in Taipei in May 2024. Here is the background information about the award: “This year, PAKDD received 720 excellent submissions, and the selection process was competitive, rigorous, and thorough with over 500 PC and 100 SPC members. An award committee was formed by a chair and four committee members from different countries. There are only one Best Paper Award, two Best Paper Running-Up Awards, and one Best Student Paper Award.” Paper Link: https://link.springer.com/chapter/10.1007/978-981-97-2253-2_11 https://pakdd2024.org/award24awardpakdd24/

[Prof. Woo, Simon Sungil] DASH Lab, Approval for AAAI 2024 publication

[Prof. Woo, Simon Sungil] DASH Lab, Approval for AAAI 2024 publication [ Paper #1 ] ※ Paper Title: All but One: Surgical Concept Erasing with Model Preservation in Text-to-Image Diffusion Models ※ paper link: https://doi.org/10.48550/arXiv.2312.12807 Seunghoo Hong†, Juhun Lee†, and Simon S. Woo*, “All but One: Surgical Concept Erasing with Model Preservation in Text-to-Image Diffusion Models”, Proceedings of the 38th annual AAAI Conference on Artificial Intelligence (AAAI), 2024. Abstract: Text-to-Image models such as Stable Diffusion have shown impressive image generation synthesis, thanks to the utilization of large-scale datasets. However, these datasets may contain sexually explicit, copyrighted, or undesirable content, which allows the model to directly generate them. Given that retraining these large models on individual concept deletion requests is infeasible, fine-tuning algorithms have been developed to tackle concept erasing in diffusion models. While these algorithms yield good concept erasure, they all present one of the following issues: 1) the corrupted feature space yields synthesis of disintegrated objects, 2) the initially synthesized content undergoes a divergence in both spatial structure and semantics in the generated images, and 3) sub-optimal training updates heighten the model's susceptibility to utility harm. These issues severely degrade the original utility of generative models. In this work, we present a new approach that solves all of these challenges. We take inspiration from the concept of classifier guidance and propose a surgical update on the classifier guidance term while constraining the drift of the unconditional score term. Furthermore, our algorithm empowers the user to select an alternative to the erasing concept, allowing for more controllability. Our experimental results show that our algorithm not only erases the target concept effectively but also preserves the model’s generation capability. [ Paper #2 ] ※ Paper Title: Layer Attack Unlearning: Fast and Accurate Machine Unlearning viaLayer Level Attack and Knowledge Distillation ※ paper link: https://arxiv.org/pdf/2312.16823.pdf Hyunjune Kim, Sangyong Lee, and Simon S. Woo*, “Layer Attack Unlearning: Fast and Accurate Machine Unlearning via Layer Level Attack and Knowledge Distillation”, Proceedings of the 38th annual AAAI Conference on Artificial Intelligence (AAAI), 2024. Abstract: Recently, serious concerns have been raised about the privacy issues related to training datasets in machine learning algorithms when including personal data. Various regulations in different countries, including the GDPR, grant individuals to have personal data erased, known as ‘the right to be forgotten’ or ‘the right to erasure’. However, there has been less research on effectively and practically deleting the requested personal data from the training set while not jeopardizing the overall machine learning performance. In this work, we propose a fast and novel machine unlearning paradigm at the layer level called layer attack unlearning, which is highly accurate and fast compared to existing machine unlearning algorithms. We introduce the Partial-PGD algorithm to locate the samples to forget efficiently. In addition, we only use the last layer of the model inspired by the Forward-Forward algorithm for unlearning process. Lastly, we use Knowledge Distillation (KD) to reliably learn the decision boundaries from the teacher using soft label information to improve accuracy performance. We conducted extensive experiments with SOTA machine unlearning models and demonstrated the effectiveness of our approach for accuracy and end-to-end unlearning performance.