Vision and Learning Pre-conference Series in Australia

NeurIPS 2021 Australia Pre-Conference

Date: Saturday, 4 December 2021 (Sydney Time, GMT+11 )

Zoom Room: https://uni-sydney.zoom.us/j/86828961270

09:20 - 09:30 Welcome speech, Prof Joachim Gudmundsson (Head of School of CS, The University of Sydney)
09:30 - 10:10 Keynote: Contrastive Graph Embeddings, Dr Piotr Koniusz (Data61/CSIRO and ANU)
10:10- 10:30 ReSSL: Relational Self-Supervised Learning with Weak Augmentation, Mingkai Zheng (The University of Sydney)
10:30 - 10:50 Meta Two-Sample Testing: Learning Kernels for Testing with Limited Data, Feng Liu (University of Technology Sydney)
10:50 - 11:10 Width-based Lookaheads with Learnt Base Policies and Heuristics Over the Atari-2600 Benchmark, Stefan O'Toole (University of Melbourne)
11:10- 11:30 Understanding and Improving Early Stopping for Learning with Noisy Labels, Yingbin Bai (The University of Sydney)
11:30- 11:50 Random Noise Defense Against Query-Based Black-Box Attacks, Zeyu Qin (The Chinese University of Hong Kong, Shenzhen)
11:50- 12:10 Neural Architecture Dilation for Adversarial Robustness, Yanxi Li (The University of Sydney)
12:10 - 14:20 Lunch Break
14:20 - 14:30 Afternoon session open
14:30 - 15:10 Keynote: Dataset representations for label-free model evaluation, Dr Liang Zheng (Australian National University)
15:10 - 15:30 MobTCast: Leveraging Auxiliary Trajectory Forecasting for Human Mobility Prediction, Hao Xue (RMIT University)
15:30 - 15:50 Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression, Jiabo He (The University of Melbourne)
15:50 - 16:10 Dynamic Neural Representational Decoders for High-Resolution Semantic Segmentation, Yifan Liu (Adelaide University)
16:10 - 16:30 Associating Objects with Transformers for Video Object Segmentation, Zongxin Yang (Zhejiang University)
16:30 - 16:50 Few-Shot Segmentation via Cycle-Consistent Transformer, Gengwei Zhang (University of Technology Sydney)
16:50 - 17:00 Closing remarks
Page last modified 30 Nov 2020.