Reading Group
Reading Group Activities
Reading group is to hold a weekly meeting where one of the members will have a 30-40 minutes talk, discussing either their own papers, papers from other research groups, or anything that they are interested in. This will be followed by a Q&A and discussion session among the group on the topic.
Membership:
Anyone can join by request. If you are interested in and have something to discuss (a title of your talk will be good, an abstract will be more than welcomed), please feel free to drop a message to satff in Trustworthy ACPS Lab.
Venue and Meeting Time:
We are now holding this online through Microsoft Teams. Starting from the year of 2020, the meeting is holded almost every Thursday 14:00-15:00, London time.
Talk Schedule:
| Time | Speaker | Topic |
|---|---|---|
| 5th Round | ||
| 14:00-15:00, Thursday, 03/11/2022 | Kaiwen Cai | Inspiring works on NeRF |
| 14:00-15:00, Thursday, 27/10/2022 | Sihao Wu | Improving Task-free Continual Learning by Distributionally Robust Memory Evolution |
| 14:00-15:00, Thursday, 20/10/2022 | Ronghui Mu | COPA: CERTIFYING ROBUST POLICIES FOR OFFLINE REINFORCEMENT LEARNING AGAINST POISONING ATTACKS |
| 14:00-15:00, Thursday, 13/10/2022 | Zhiwei Shan | Learning to coloring: DRL method based on GNN |
| 14:00-15:00, Thursday, 06/10/2022 | Dengyu Wu | Adaptive Spiking Neural Network |
| 14:00-15:00, Thursday, 29/09/2022 | Chi Zhang | Certifiably Adversarially Robust Detection of Out-of-Distribution Data |
| 5th Round | ||
| 4th Round | ||
| 14:00-15:00, Thursday, 22/09/2022 | Sahar Alzahrani | Formal analysis of Deep Binarized Neural Networks |
| 14:00-15:00, Thursday, 15/09/2022 | Yi Qi | Delving Deep Into Label Smoothing |
| 14:00-15:00, Thursday, 08/09/2022 | Alshareef Amany | Weight-based Semantic Testing Approach for Neural Network |
| 14:00-15:00, Thursday, 01/09/2022 | Fu Wang | Adversarial robustness via robust low rank representations |
| 14:00-15:00, Thursday, 25/08/2022 | Siqi Sun | SAFER: A Structure-free Approach for Certified Robustness to Adversarial Word Substitutions |
| 14:00-15:00, Thursday, 18/08/2022 | Tianle Zhang | Generalisation to Flat\Sharp Minima |
| 14:00-15:00, Thursday, 11/08/2022 | Peipei Xu | Certified Patch Robustness Via Smoothed Vision Transformers |
| 14:00-15:00, Thursday, 04/08/2022 | Xiangyu Yin | The Limitations of Large Width in Neural Networks: A Deep Gaussian Process Perspective |
| 14:00-15:00, Thursday, 28/07/2022 | Zheng Wang | Understanding Adversarial Robustness of Vision Transformers via ODEs |
| 14:00-15:00, Thursday, 21/07/2022 | Jiaxu Liu | Feature and Weight Regularization for Deeper Graph Neural Networks |
| 14:00-15:00, Thursday, 14/07/2022 | Kaiwen Cai | How Metric Learning improves Classification and Regression |
| 14:00-15:00, Thursday, 07/07/2022 | Gaojie Jin | Robustness and Accuracy Could be Reconcilable By (Proper) Definition |
| 14:00-15:00, Thursday, 30/06/2022 | Sihao Wu | Data Augmentation of Reinforcement Learning |
| 14:00-15:00, Thursday, 23/06/2022 | Fan Zhou | Learning Transferable Features from Different Domains |
| 14:00-15:00, Thursday, 16/06/2022 | Han Yu | Cache-based GNN System for Dynamic Graphs |
| 14:00-15:00, Thursday, 09/06/2022 | Ronghui Mu | POLICY SMOOTHING FOR PROVABLY ROBUST REINFORCEMENT LEARNING |
| 14:00-15:00, Thursday, 19/05/2022 | Zhiwei Shan | Learning What to Defer for Maximum Independent Sets |
| 14:00-15:00, Thursday, 05/05/2022 | Dengyu Wu | Comparison between Quantised and Spiking Neural Network |
| 14:00-15:00, Thursday, 28/04/2022 | Han Wu | Man-in-the-Middle Attack against Object Detection |
| 14:00-15:00, Thursday, 28/04/2022 | Wei Huang | A Statistical Approach to Assessing Neural Network Robustness |
| 14:00-15:00, Thursday, 21/04/2022 | Chi Zhang | Node Similarity Preserving Graph Convolutional Networks |
| 14:00-15:00, Thursday, 07/04/2022 | Ganlin Liu | Node Similarity Preserving Graph Convolutional Networks |
| 4th Round | ||
| 3rd Round | ||
| 14:00-15:00, Thursday, 31/03/2022 | Yi Qi | A Vision-based Irregular Obstacle Avoidance Framework via Deep Reinforcement Learning |
| 14:00-15:00, Thursday, 24/03/2022 | Amany Alshareef | Distribution-Aware Testing of Neural Networks Using Generative Models |
| 14:00-15:00, Thursday, 17/03/2022 | Fu Wang | Adversarial Robustness Across Representation Spaces |
| 14:00-15:00, Thursday, 03/03/2022 | Siqi Sun | SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimization |
| 14:00-15:00, Thursday, 24/02/2022 | Yanda Meng | Consistency Regularisation Enabled Semi-supervised Learning in Task of Medical Image Analysis |
| 14:00-15:00, Thursday, 17/02/2022 | Tianle Zhang | Probabilistically Robust Learning: Balancing Average- and Worst-case Performance |
| 14:00-15:00, Thursday, 10/02/2022 | Peipei Xu | Formal Security Analysis of Neural Networks using Symbolic Intervals |
| 14:00-15:00, Thursday, 03/02/2022 | Zheng Wang | Intriguing Properties of Vision Transformer |
| 14:00-15:00, Thursday, 27/01/2022 | Xiangyu Yin | Appending Adversarial Frames for Universal Video Attack |
| 14:00-15:00, Thursday, 20/01/2022 | Jiaxu Liu | Normalizing Flow |
| 14:00-15:00, Thursday, 13/01/2022 | Kaiwen Cai | Transfer Learning for Point Clouds |
| 14:00-15:00, Thursday, 06/01/2022 | Sihao Wu | Stealthy and Efficient Adversarial Attacks against Deep Reinforcement Learning |
| 14:00-15:00, Thursday, 16/12/2021 | Ronghui Mu | Certified Adversarial Robustness for Deep Reinforcement Learning |
| 14:00-15:00, Thursday, 09/12/2021 | Han Yu | Link scheduling using graph neural networks |
| 14:00-15:00, Thursday, 02/12/2021 | Zhiwei Shan | Exact Combinatorial Optimization with Graph Convolutional Neural Networks |
| 14:00-15:00, Thursday, 25/11/2021 | Dengyu Wu | Exploring Adversarial Attack in Spiking Neural Networks With Spike-Compatible Gradient |
| 14:00-15:00, Thursday, 18/11/2021 | Han Wu | Black-box Adversarial Toolbox (BAT): Distributed black-box attack |
| 14:00-15:00, Thursday, 11/11/2021 | Wei Huang | Towards Feature Space Adversarial Attack by Style Perturbation |
| 14:00-15:00, Thursday, 04/11/2021 | Yanghao Zhang | The Dimpled Manifold Model of Adversarial Examples in Machine Learning |
| 14:00-15:00, Thursday, 28/10/2021 | Gaojie jin | Patches Are All You Need? |
| 14:00-15:00, Thursday, 21/10/2021 | Chi Zhang | Verisig 2.0: Verification of Neural Network Controllers Using Taylor Model Preconditioning |
| 14:00-15:00, Thursday, 14/10/2021 | Ganlin Liu | Predict Then Propagate: Graph Neural Networks Meet Personalized PageRank’ |
| 3rd Round | ||
| 2nd Round | ||
| 14:00-15:00, Thursday, 07/10/2021 | Jianqiao Zhang | Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods |
| 14:00-15:00, Thursday, 30/09/2021 | Amany Alshareef | Bayesian Deep Learning via Subnetwork Inference |
| 14:00-15:00, Thursday, 23/09/2021 | Fu Wang | Certify or Predict: Boosting Certified Robustness with Compositional Architectures |
| 14:00-15:00, Thursday, 16/09/2021 | Emese Thomas | Enforcing Interpretability and its Statistical Impacts: Trade-offs between Accuracy and Interpretability |
| 14:00-15:00, Thursday, 09/09/2021 | Siqi Sun | Towards Robustness Against Natural Language Word Substitutions |
| 14:00-15:00, Thursday, 02/09/2021 | Yanda Meng | Regression of Instance Boundary by Aggregated CNN and GCN |
| 14:00-15:00, Thursday, 26/08/2021 | Tianle Zhang | Attribute-Guided Adversarial Training for Robustness to Natural Perturbations |
| 14:00-15:00, Thursday, 19/08/2021 | Peipei Xu | Towards verifying robustness of neural networks against a family of semantic perturbations |
| 14:00-15:00, Thursday, 12/08/2021 | Zheng Wang | On the Adversarial Robustness of Visual Transformers |
| 14:00-15:00, Thursday, 05/08/2021 | Dengyu Wu | Towards real-time photorealistic 3D holography with deep neural networks |
| 14:00-15:00, Thursday, 29/07/2021 | Ronghui Mu | Reducing DNN Properties to Enable Falsification with Adversarial Attacks |
| 14:00-15:00, Thursday, 22/07/2021 | Ganlin Liu | Heterogeneous Graph Matching Networks for Unknown Malware Detection |
| 14:00-15:00, Thursday, 15/07/2021 | Han Yu | Distributed scheduling using graph neural networks |
| 14:00-15:00, Thursday, 08/07/2021 | Kaiwen Cai | Depth estimation from monocular images and sparse radar data |
| 14:00-15:00, Thursday, 01/07/2021 | Han Wu | Adversarial Detection: Generating adversarial patch is as easy as drag and drop |
| 14:00-15:00, Thursday, 24/06/2021 | Wei Huang | RobOT: Robustness-Oriented Testing for Deep Learning Systems |
| 14:00-15:00, Thursday, 17/06/2021 | Yanghao Zhang | Robust Overfitting may be mitigated by properly learned smoothening |
| 14:00-15:00, Thursday, 10/06/2021 | Gaojie Jin | Learning Diverse and Discriminative Representations via the Principle of Maximal Coding Rate Reduction |
| 14:00-15:00, Thursday, 03/06/2021 | Chi Zhang | Learning to Simulate Complex Physics with Graph Networks |
| 14:00-15:00, Thursday, 27/05/2021 | Zhiwei Shan | How powerful are graph neural networks? |
| 2nd Round | ||
| 1st Round | ||
| 14:00-15:00, Thursday, 20/05/2021 | Amany Alshareef | Robust feature learning for adversarial defense via hierarchical feature alignment |
| 14:00-15:00, Thursday, 13/05/2021 | Jianqiao Zhang | Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples |
| 14:00-15:00, Thursday, 06/05/2021 | Fu Wang | Gradient-Guided Dynamic Efficient Adversarial Training |
| 14:00-15:00, Thursday, 29/04/2021 | Emese Thamo | Benchmarking and Survey of Explanation Methods for Black Box Models |
| 14:00-15:00, Thursday, 22/04/2021 | Siqi Sun | HopSkipJumpAttack: A Query-Efficient Decision-Based Attack |
| 14:00-15:00, Thursday, 15/04/2021 | Yanda Meng | An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale |
| 14:00-15:00, Thursday, 08/04/2021 | Tianle Zhang | Evidential deep learning for uncertainty estimation |
| 14:00-15:00, Thursday, 01/04/2021 | Peipei Xu | Boosting robustness certification of neural networks |
| 14:00-15:00, Thursday, 25/03/2021 | Zheng Wang | Fundamental Tradeoffs between Invariance and Sensitivity to Adversarial Perturbations |
| 14:00-15:00, Thursday, 18/03/2021 | Kaiwen Cai | Particle Filter Recurrent Neural Networks |
| 14:00-15:00, Thursday, 11/03/2021 | Ronghui Mu | Certifying Robustness of Convolutional Neural Networks |
| 14:00-15:00, Thursday, 04/03/2021 | Dengyu Wu | Adversarial Robustness of Spiking Neural Networks |
| 14:00-15:00, Thursday, 25/02/2021 | Han Wu | From MODEL to MOSFET: Why models fail to survive on edge devices? |
| 14:00-15:00, Thursday, 18/02/2021 | Gaojie Jin | Dropout Interpretability |
| 14:00-15:00, Thursday, 04/02/2021 | Yanghao Zhang | Breaking Certified Defenses and Verifier |
| 14:00-15:00, Thursday, 28/01/2021 | Wei Huang | Reliability Assessment Model for DL system through Robustness Evaluation and Operational Testing |
| 14:00-15:00, Thursday, 21/01/2021 | Chi Zhang | Verifying Recurrent Neural Networks using Invariant Inference |
| 14:00-15:00, Thursday, 14/01/2021 | Han Yu | Auto-Encoding Variational Bayes |
| 11:00-12:00, Tuesday, 15/12/2020 | Ronghui Mu | Generating 3D Adversarial Point Clouds |
| 11:00-12:00, Tuesday, 08/12/2020 | Yanda Meng | Uncertainty-Aware Semi-Supervised Learning |
| 11:00-12:00, Tuesday, 01/12/2020 | Peipei XU | "(De)Randomized Smoothing for Certifiable Defense against Patch Attacksn" |
| 11:00-12:00, Tuesday, 24/11/2020 | Kaiwen Cai | "NetVLAD: CNN architecture for weakly supervised place recognition" |
| 11:00-12:00, Tuesday, 17/11/2020 | Dengyu WU | Energy Efficient Spiking Neural Networks |
| 11:00-12:00, Tuesday, 10/11/2020 | Wei Huang | Boosting Operational DNN Testing Efficiency through Conditioning |
| 11:00-12:00, Tuesday, 03/11/2020 | Yuyao Kong, Southeast University | In-memory Processing based on Time-domain Circuit |
| 11:00-12:00, Tuesday, 27/10/2020 | Gaojie Jin | How does Weight Correlation Affect the Generalisation Ability of Deep Neural Networks |
| 11:00-12:00, Tuesday, 20/10/2020. | Han Yu | Rate-Distortion Optimization Guided Autoencoder for Isometric Embedding in Euclidean Latent Space |
| 11:00-12:00, Tuesday, 13/10/2020 | Chi Zhang, University of Exeter | POPQORN: Quantifying Robustness of Recurrent Neural Networks |
| 11:00-12:00, Tuesday, 06/10/2020 | Ronghui Mu, University of Exeter | Sparse Adversarial Perturbations for Videos |
| 11:00-12:00, Tuesday, 29/09/2020 | Guanxiong Shen | Radio Frequency Fingerprint Identification for LoRa Using Deep Learning |
| 11:00-12:00, Tuesday, 22/09/2020 | Dengyu Wu | ‘introduction to Spiking Neural Networks’ |
| 11:00-12:00, Tuesday, 15/09/2020. | Peipei Xu. | ‘Certified Defenses for Adversarial Patches’ |
| 11:00-12:00, Tuesday, 08/09/2020 | Fu Wang, University of Exeter | ‘Adversarial Training for Free!’ and ‘Fast is better than free: Revisiting adversarial training’ |
| 11:00-12:00, Tuesday, 01/09/2020 | Wei Huang | Practical Verification of Neural Network Enabled State Estimation System for Robotics |
| 11:00-12:00, Tuesday, 25/08/2020 | Gaojie Jin | `The Implicit and Explicit Regularization Effects of Dropout' |
| 16:00-17:00, Tuesday, 18/08/2020 | Xiaowei Huang | Safety Certification of Deep Learning |
| 16:00-17:00, Tuesday, 11/08/2020 | Han Yu | Rate-Distortion Optimization Guided Autoencoder for Isometric Embedding in Euclidean Latent Space |
| 16:00-17:00, Tuesday, 04/08/2020 | Dengyu Wu | Introduction to Spiking Neural Networks |
| 16:00-17:00, Tuesday, 28/07/2020 | Peipei Xu. | Certified Adversarial Robustness via Randomized Smoothing |
| 16:00-17:00, Tuesday, 21/07/2020. | Wei Huang. | backdoor attack on tree ensembles |
| 16:00-17:00, Tuesday, 14/07/2020 | Gaojie Jin | predicting generalisation of deep neural networks |
| 16:00-17:00, Tuesday, 07/07/2020. | Xinping Yi. | spectral analysis of convolutional neural networks |
| 1st Round | ||