no code implementations • 15 Apr 2024 • Nawrin Tabassum, Ka-Ho Chow, Xuyu Wang, Wenbin Zhang, Yanzhao Wu
Second, we propose three early-stopping techniques to effectively reduce the computational costs of these privacy attacks.
no code implementations • 30 Jan 2024 • Badhan Chandra Das, M. Hadi Amini, Yanzhao Wu
We assess the extent of LLM vulnerabilities, investigate emerging security and privacy attacks for LLMs, and review the potential defense mechanisms.
1 code implementation • 17 Nov 2023 • Yanzhao Wu, Ka-Ho Chow, Wenqi Wei, Ling Liu
This paper presents a novel deep ensemble pruning approach, which can efficiently identify smaller deep ensembles and provide higher ensemble accuracy than the entire deep ensemble of a large number of member networks.
1 code implementation • 11 Nov 2023 • Badhan Chandra Das, M. Hadi Amini, Yanzhao Wu
Federated learning (FL) is gaining increasing popularity in the medical domain for analyzing medical images, which is considered an effective technique to safeguard sensitive patient data and comply with privacy regulations.
1 code implementation • 3 Oct 2023 • Yanzhao Wu, Ka-Ho Chow, Wenqi Wei, Ling Liu
We show that this two-tier heterogeneity driven ensemble construction method can compose an ensemble team that promotes high ensemble diversity and low negative correlation among member models of the ensemble, strengthening ensemble robustness against both negative examples and adversarial attacks.
2 code implementations • 22 Sep 2023 • Xirong Cao, Xiang Li, Divyesh Jadav, Yanzhao Wu, Zhehui Chen, Chen Zeng, Wenqi Wei
Diffusion models have gained prominence in the image domain for their capabilities in data generation and transformation, achieving state-of-the-art performance in various tasks in both image and audio domains.
1 code implementation • 16 Sep 2023 • Hongpeng Jin, Wenqi Wei, Xuyu Wang, Wenbin Zhang, Yanzhao Wu
Second, we present LRBench++ to benchmark learning rate policies and facilitate learning rate tuning for both traditional DNNs and LLMs.
1 code implementation • 6 Sep 2023 • Sanjana Vijay Ganesh, Yanzhao Wu, Gaowen Liu, Ramana Kompella, Ling Liu
Object tracking is an important functionality of edge video analytic systems and services.
1 code implementation • 10 May 2023 • Wenqi Wei, Ling Liu, Jingya Zhou, Ka-Ho Chow, Yanzhao Wu
Next, we present a gradient leakage resilient approach to securing distributed SGD in federated learning, with differential privacy controlled noise as the tool.
1 code implementation • CVPR 2023 • Ka-Ho Chow, Ling Liu, Wenqi Wei, Fatih Ilhan, Yanzhao Wu
Based on the insights, we introduce a three-tier forensic framework to identify and expel Trojaned gradients and reclaim the performance over the course of FL.
1 code implementation • 15 Jan 2023 • Fatih Ilhan, Ka-Ho Chow, Sihao Hu, Tiansheng Huang, Selim Tekin, Wenqi Wei, Yanzhao Wu, Myungjin Lee, Ramana Kompella, Hugo Latapie, Gaowen Liu, Ling Liu
Instead of having every sample go through all DNN layers during prediction, EENet learns an early exit scheduler, which can intelligently terminate the inference earlier for certain predictions, which the model has high confidence of early exit.
1 code implementation • 24 Oct 2022 • Yanzhao Wu, Ling Liu
First, we develop an LR tuning mechanism for auto-verification of a given LR policy with respect to the desired accuracy goal under the pre-defined training time constraint.
1 code implementation • IEEE International Conference on Data Mining (ICDM) 2021 • Yanzhao Wu, Ling Liu
Evaluated using two benchmark datasets, we show that the proposed focal diversity powered hierarchical pruning can find significantly smaller ensembles of deep neural network models while achieving the same or better classification generalizability.
1 code implementation • 22 Oct 2021 • Zhongwei Xie, Ling Liu, Yanzhao Wu, Luo Zhong, Lin Li
This paper introduces a two-phase deep feature engineering framework for efficient learning of semantics enhanced joint embedding, which clearly separates the deep feature engineering in data preprocessing from training the text-image joint embedding model.
1 code implementation • 2 Aug 2021 • Zhongwei Xie, Ling Liu, Yanzhao Wu, Lin Li, Luo Zhong
We present a Multi-modal Semantics enhanced Joint Embedding approach (MSJE) for learning a common feature space between the two modalities (text and image), with the ultimate goal of providing high-performance cross-modal retrieval services.
1 code implementation • 27 Jul 2021 • Yanzhao Wu, Ling Liu, Ramana Kompella
A common performance requirement in these mission-critical edge services is the near real-time latency of online object detection on edge devices.
1 code implementation • 2 Jul 2021 • Wenqi Wei, Ling Liu, Yanzhao Wu, Gong Su, Arun Iyengar
This paper presents a gradient leakage resilient approach to privacy-preserving federated learning with per training example-based client differential privacy, coined as Fed-CDP.
1 code implementation • CVPR 2021 • Yanzhao Wu, Ling Liu, Zhongwei Xie, Ka-Ho Chow, Wenqi Wei
Our new metrics significantly improve the intrinsic correlation between high ensemble diversity and high ensemble accuracy.
1 code implementation • 1 Jan 2021 • Yanzhao Wu, Ling Liu
(3) We introduce a two phase hierarchical pruning method to effectively identify and prune those deep ensembles with high HQ diversity scores, aiming to increase the lower and upper bounds on ensemble accuracy for the selected ensembles.
1 code implementation • 20 Oct 2020 • Yanzhao Wu, Ling Liu, Zhongwei Xie, Juhyun Bae, Ka-Ho Chow, Wenqi Wei
Ensemble learning is gaining renewed interests in recent years.
1 code implementation • 11 Jul 2020 • Ka-Ho Chow, Ling Liu, Mehmet Emre Gursoy, Stacey Truex, Wenqi Wei, Yanzhao Wu
We demonstrate that the proposed framework can serve as a methodical benchmark for analyzing adversarial behaviors and risks in real-time object detection systems.
2 code implementations • 22 Apr 2020 • Wenqi Wei, Ling Liu, Margaret Loper, Ka-Ho Chow, Mehmet Emre Gursoy, Stacey Truex, Yanzhao Wu
FL offers default client privacy by allowing clients to keep their sensitive data on local devices and to only share local training parameter updates with the federated server.
2 code implementations • 9 Apr 2020 • Ka-Ho Chow, Ling Liu, Mehmet Emre Gursoy, Stacey Truex, Wenqi Wei, Yanzhao Wu
The rapid growth of real-time huge data capturing has pushed the deep learning and data analytic computing to the edge systems.
no code implementations • 1 Oct 2019 • Wenqi Wei, Ling Liu, Margaret Loper, Ka-Ho Chow, Emre Gursoy, Stacey Truex, Yanzhao Wu
Deep neural network (DNN) has demonstrated its success in multiple domains.
no code implementations • 29 Aug 2019 • Ling Liu, Wenqi Wei, Ka-Ho Chow, Margaret Loper, Emre Gursoy, Stacey Truex, Yanzhao Wu
In this paper we first give an overview of the concept of ensemble diversity and examine the three types of ensemble diversity in the context of DNN classifiers.
no code implementations • 21 Aug 2019 • Ka-Ho Chow, Wenqi Wei, Yanzhao Wu, Ling Liu
Deep neural networks (DNNs) have demonstrated impressive performance on many challenging machine learning tasks.
1 code implementation • 18 Aug 2019 • Yanzhao Wu, Ling Liu, Juhyun Bae, Ka-Ho Chow, Arun Iyengar, Calton Pu, Wenqi Wei, Lei Yu, Qi Zhang
Learning Rate (LR) is an important hyper-parameter to tune for effective training of deep neural networks (DNNs).
1 code implementation • 29 Oct 2018 • Yanzhao Wu, Ling Liu, Calton Pu, Wenqi Cao, Semih Sahin, Wenqi Wei, Qi Zhang
Big data powered Deep Learning (DL) and its applications have blossomed in recent years, fueled by three technological trends: a large amount of digitized data openly accessible, a growing number of DL software frameworks in open source and commercial markets, and a selection of affordable parallel computing hardware devices.
no code implementations • 29 Jun 2018 • Wenqi Wei, Ling Liu, Margaret Loper, Stacey Truex, Lei Yu, Mehmet Emre Gursoy, Yanzhao Wu
The burgeoning success of deep learning has raised the security and privacy concerns as more and more tasks are accompanied with sensitive data.