10/10/2024, 9am UK time, 4pm Beijing time
Adversarial Data Distortion and Impact on Deep Vision Perception for ADS/ADAS
Zixiang Wei (University of Warwick)
Reading group
Ideas examined in the open.
From 2024, the tutorial reading group holds bi-weekly meetings featuring a detailed three-hour tutorial talk presented by one of our group members, discussing state-of-the-art research topics that they are interested in. Each presentation is followed by a Q&A and discussion session on the topic.
Anyone can join by request. If you are interested and have something to discuss, please feel free to get in touch with staff in Trustworthy ACPS Lab.
Tutorial sessions are held online through Microsoft Teams. The second round was scheduled for 9am UK time and 4pm Beijing time.
Tutorial series
Recent and scheduled sessions.
23 sessions
24/10/2024, 9am UK time, 4pm Beijing time
Tutorial 2
Meilin Zhu (Chinese Academy of Sciences)
07/11/2024, 9am UK time, 4pm Beijing time
Tutorial 3
Jiaxu Liu (The University of Liverpool)
21/11/2024, 9am UK time, 4pm Beijing time
Tutorial 4
Shuaitong Liu (Chinese Academy of Sciences)
05/12/2024, 9am UK time, 4pm Beijing time
Tutorial 5
Xinmiao Huang
19/12/2024, 9am UK time, 4pm Beijing time
Tutorial 6
Yi Zhang (University of Warwick)
02/01/2025, 9am UK time, 4pm Beijing time
Tutorial 7
Jinyong Wang (Visitor at The University of Liverpool)
16/01/2025, 9am UK time, 4pm Beijing time
Tutorial 8
Zhenglin Huang (The University of Liverpool)
06/02/2025, 9am UK time, 4pm Beijing time
Tutorial 9
Zhen Zhang (Chinese Academy of Sciences)
20/02/2025, 9am UK time, 4pm Beijing time
Tutorial 10
Zhuoyun Li (The University of Liverpool)
06/03/2025, 9am UK time, 4pm Beijing time
Tutorial 11
Zixia Wang (University of Exeter)
20/03/2025, 9am UK time, 4pm Beijing time
Tutorial 12
Qisong He (The University of Liverpool)
03/04/2025, 9am UK time, 4pm Beijing time
Tutorial 13
Yuhang Chen (University of Warwick)
17/04/2025, 9am UK time, 4pm Beijing time
Tutorial 14
Zhen Chen (The University of Liverpool)
24/04/2025, 9am UK time, 4pm Beijing time
Tutorial 14
Jiacheng Pi (The University of Science and Technology of China)
08/05/2025, 9am UK time, 4pm Beijing time
Tutorial 15
Ruizhi Zhong (The University of Science and Technology of China)
13:00-16:30, Thursday, 18/April/2024
Membership Inference Attacks and Differential Privacy
Zhen Chen (University of Liverpool)
13:00-16:00, Thursday, 02/May/2024
Empowering Autonomous Driving with LLM and VLM
Sihao Wu (University of Liverpool)
13:00-16:00, Thursday, 16/May/2024
Deep Model Complexity and Its Applications in Adversarial Learning
Xiangyu Yin (University of Liverpool)
13:00-16:00, Thursday, 30/May/2024
How to Control LLMs’ behaviors and Design Strategy to safeguard LLMs
Jinwei Hu (University of Liverpool)
13:00-16:00, Thursday, 20/June/2024
Exploring Black-Box Global Optimization: From Multi-Arm Bandits to Bayesian Techniques
Fu Wang (University of Exeter)
13:00-16:00, Thursday, 04/July/2024
Exploring Prompt Engineering to Retrieval-Augmented Generation and AI Agents
Yi Qi (University of Liverpool)
13:00-16:00, Thursday, 25/July/2024
Robustness Verification of Text-to-Image Diffusion Models against Stochastic Perturbation
Yi Zhang (University of Warwick)
Archive
Earlier discussions.
112 sessions
202240 sessions
- Inspiring works on NeRFKaiwen Cai
- Improving Task-free Continual Learning by Distributionally Robust Memory EvolutionSihao Wu
- COPA: CERTIFYING ROBUST POLICIES FOR OFFLINE REINFORCEMENT LEARNING AGAINST POISONING ATTACKSRonghui Mu
- Learning to coloring: DRL method based on GNNZhiwei Shan
- Adaptive Spiking Neural NetworkDengyu Wu
- Certifiably Adversarially Robust Detection of Out-of-Distribution DataChi Zhang
- Formal analysis of Deep Binarized Neural NetworksSahar Alzahrani
- Delving Deep Into Label SmoothingYi Qi
- Weight-based Semantic Testing Approach for Neural NetworkAlshareef Amany
- Adversarial robustness via robust low rank representationsFu Wang
- SAFER: A Structure-free Approach for Certified Robustness to Adversarial Word SubstitutionsSiqi Sun
- Generalisation to Flat\Sharp MinimaTianle Zhang
- Certified Patch Robustness Via Smoothed Vision TransformersPeipei Xu
- The Limitations of Large Width in Neural Networks: A Deep Gaussian Process PerspectiveXiangyu Yin
- Understanding Adversarial Robustness of Vision Transformers via ODEsZheng Wang
- Feature and Weight Regularization for Deeper Graph Neural NetworksJiaxu Liu
- How Metric Learning improves Classification and RegressionKaiwen Cai
- Robustness and Accuracy Could be Reconcilable By (Proper) DefinitionGaojie Jin
- Data Augmentation of Reinforcement LearningSihao Wu
- Learning Transferable Features from Different DomainsFan Zhou
- Cache-based GNN System for Dynamic GraphsHan Yu
- POLICY SMOOTHING FOR PROVABLY ROBUST REINFORCEMENT LEARNINGRonghui Mu
- Learning What to Defer for Maximum Independent SetsZhiwei Shan
- Comparison between Quantised and Spiking Neural NetworkDengyu Wu
- Man-in-the-Middle Attack against Object DetectionHan Wu
- A Statistical Approach to Assessing Neural Network RobustnessWei Huang
- Node Similarity Preserving Graph Convolutional NetworksChi Zhang
- Node Similarity Preserving Graph Convolutional NetworksGanlin Liu
- A Vision-based Irregular Obstacle Avoidance Framework via Deep Reinforcement LearningYi Qi
- Distribution-Aware Testing of Neural Networks Using Generative ModelsAmany Alshareef
- Adversarial Robustness Across Representation SpacesFu Wang
- SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized OptimizationSiqi Sun
- Consistency Regularisation Enabled Semi-supervised Learning in Task of Medical Image AnalysisYanda Meng
- Probabilistically Robust Learning: Balancing Average- and Worst-case PerformanceTianle Zhang
- Formal Security Analysis of Neural Networks using Symbolic IntervalsPeipei Xu
- Intriguing Properties of Vision TransformerZheng Wang
- Appending Adversarial Frames for Universal Video AttackXiangyu Yin
- Normalizing FlowJiaxu Liu
- Transfer Learning for Point CloudsKaiwen Cai
- Stealthy and Efficient Adversarial Attacks against Deep Reinforcement LearningSihao Wu
202148 sessions
- Certified Adversarial Robustness for Deep Reinforcement LearningRonghui Mu
- Link scheduling using graph neural networksHan Yu
- Exact Combinatorial Optimization with Graph Convolutional Neural NetworksZhiwei Shan
- Exploring Adversarial Attack in Spiking Neural Networks With Spike-Compatible GradientDengyu Wu
- Black-box Adversarial Toolbox (BAT): Distributed black-box attackHan Wu
- Towards Feature Space Adversarial Attack by Style PerturbationWei Huang
- The Dimpled Manifold Model of Adversarial Examples in Machine LearningYanghao Zhang
- Patches Are All You Need?Gaojie jin
- Verisig 2.0: Verification of Neural Network Controllers Using Taylor Model PreconditioningChi Zhang
- Predict Then Propagate: Graph Neural Networks Meet Personalized PageRank’Ganlin Liu
- Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation MethodsJianqiao Zhang
- Bayesian Deep Learning via Subnetwork InferenceAmany Alshareef
- Certify or Predict: Boosting Certified Robustness with Compositional ArchitecturesFu Wang
- Enforcing Interpretability and its Statistical Impacts: Trade-offs between Accuracy and InterpretabilityEmese Thomas
- Towards Robustness Against Natural Language Word SubstitutionsSiqi Sun
- Regression of Instance Boundary by Aggregated CNN and GCNYanda Meng
- Attribute-Guided Adversarial Training for Robustness to Natural PerturbationsTianle Zhang
- Towards verifying robustness of neural networks against a family of semantic perturbationsPeipei Xu
- On the Adversarial Robustness of Visual TransformersZheng Wang
- Towards real-time photorealistic 3D holography with deep neural networksDengyu Wu
- Reducing DNN Properties to Enable Falsification with Adversarial AttacksRonghui Mu
- Heterogeneous Graph Matching Networks for Unknown Malware DetectionGanlin Liu
- Distributed scheduling using graph neural networksHan Yu
- Depth estimation from monocular images and sparse radar dataKaiwen Cai
- Adversarial Detection: Generating adversarial patch is as easy as drag and dropHan Wu
- RobOT: Robustness-Oriented Testing for Deep Learning SystemsWei Huang
- Robust Overfitting may be mitigated by properly learned smootheningYanghao Zhang
- Learning Diverse and Discriminative Representations via the Principle of Maximal Coding Rate ReductionGaojie Jin
- Learning to Simulate Complex Physics with Graph NetworksChi Zhang
- How powerful are graph neural networks?Zhiwei Shan
- Robust feature learning for adversarial defense via hierarchical feature alignmentAmany Alshareef
- Towards Interpretable Deep Neural Networks by Leveraging Adversarial ExamplesJianqiao Zhang
- Gradient-Guided Dynamic Efficient Adversarial TrainingFu Wang
- Benchmarking and Survey of Explanation Methods for Black Box ModelsEmese Thamo
- HopSkipJumpAttack: A Query-Efficient Decision-Based AttackSiqi Sun
- An Image is Worth 16x16 Words: Transformers for Image Recognition at ScaleYanda Meng
- Evidential deep learning for uncertainty estimationTianle Zhang
- Boosting robustness certification of neural networksPeipei Xu
- Fundamental Tradeoffs between Invariance and Sensitivity to Adversarial PerturbationsZheng Wang
- Particle Filter Recurrent Neural NetworksKaiwen Cai
- Certifying Robustness of Convolutional Neural NetworksRonghui Mu
- Adversarial Robustness of Spiking Neural NetworksDengyu Wu
- From MODEL to MOSFET: Why models fail to survive on edge devices?Han Wu
- Dropout InterpretabilityGaojie Jin
- Breaking Certified Defenses and VerifierYanghao Zhang
- Reliability Assessment Model for DL system through Robustness Evaluation and Operational TestingWei Huang
- Verifying Recurrent Neural Networks using Invariant InferenceChi Zhang
- Auto-Encoding Variational BayesHan Yu
202024 sessions
- Generating 3D Adversarial Point CloudsRonghui Mu
- Uncertainty-Aware Semi-Supervised LearningYanda Meng
- "(De)Randomized Smoothing for Certifiable Defense against Patch Attacksn"Peipei XU
- "NetVLAD: CNN architecture for weakly supervised place recognition"Kaiwen Cai
- Energy Efficient Spiking Neural NetworksDengyu WU
- Boosting Operational DNN Testing Efficiency through ConditioningWei Huang
- In-memory Processing based on Time-domain CircuitYuyao Kong, Southeast University
- How does Weight Correlation Affect the Generalisation Ability of Deep Neural NetworksGaojie Jin
- Rate-Distortion Optimization Guided Autoencoder for Isometric Embedding in Euclidean Latent SpaceHan Yu
- POPQORN: Quantifying Robustness of Recurrent Neural NetworksChi Zhang, University of Exeter
- Sparse Adversarial Perturbations for VideosRonghui Mu, University of Exeter
- Radio Frequency Fingerprint Identification for LoRa Using Deep LearningGuanxiong Shen
- ‘introduction to Spiking Neural Networks’Dengyu Wu
- ‘Certified Defenses for Adversarial Patches’Peipei Xu.
- ‘Adversarial Training for Free!’ and ‘Fast is better than free: Revisiting adversarial training’Fu Wang, University of Exeter
- Practical Verification of Neural Network Enabled State Estimation System for RoboticsWei Huang
- `The Implicit and Explicit Regularization Effects of Dropout'Gaojie Jin
- Safety Certification of Deep LearningXiaowei Huang
- Rate-Distortion Optimization Guided Autoencoder for Isometric Embedding in Euclidean Latent SpaceHan Yu
- Introduction to Spiking Neural NetworksDengyu Wu
- Certified Adversarial Robustness via Randomized SmoothingPeipei Xu.
- backdoor attack on tree ensemblesWei Huang.
- predicting generalisation of deep neural networksGaojie Jin
- spectral analysis of convolutional neural networksXinping Yi.