We were pleased to present 6 papers from our group .
Our advanced research lab carries out fascinating research at the intersection of neuroscience and AI. We develop neuro-inspired AI models that can learn efficiently from multiple tasks.#CoLLAs2022
Conversation
(1/6) Task Agnostic Representation Consolidation: a Self-supervised based Continual Learning Approach
Paper: arxiv.org/abs/2207.06267
Video: youtube.com/watch?v=ECFMKv
#ContinualLearning #SelfSuppervisedLearning
1
3
(2/6) InBiaseD: Inductive Bias Distillation to Improve Generalization and Robustness through Shape-awareness
Paper: arxiv.org/abs/2206.05846
Video: youtube.com/watch?v=wKbKuE
#DeepLearning #FairAI
1
3
(3/6) SYNERgy between SYNaptic consolidation and Experience Replay for general continual learning
Paper: arxiv.org/abs/2206.04016
Video: youtube.com/watch?v=lCBra-
#ContinualLearning #BrainInspiredAI
1
3
(4/6) Differencing based Self-supervised pretraining for Scene Change Detection
Paper: arxiv.org/abs/2208.05838
Video: youtube.com/watch?v=kWUxxC
#DeepLearning #ChangeDetection
1
3
(5/6) Consistency is the key to further mitigating catastrophic forgetting in continual learning
Paper: arxiv.org/abs/2207.04998
Video: youtube.com/watch?v=7-lfLf
#ContinualLearning #DeepLearning
Replying to
(6/6) Curbing Task Interference using Representation Similarity-Guided Multi-Task Feature Sharing
Video: youtube.com/watch?v=9OdpKw
#MultiTaskLearning #DeepLearning
3