Vector Institute Researchers Present Latest AI Findings at ICLR 2018
The Vector Institute for Artificial Intelligence announced that its faculty and researchers will attend the International Conference on Learning Representations (ICLR) to present their latest advancements in machine learning. The institute highlighted a comprehensive list of accepted papers and workshop contributions scheduled for the conference in late April and early May 2018. Key research topics include Kronecker-factored curvature approximations for recurrent neural networks, quantitative evaluation of Generative Adversarial Networks (GANs), and efficient weight perturbations via Flipout. Other significant works address attacking binarized neural networks, meta-learning for few-shot classification, and optimizing control variates for black-box gradient estimation. Additionally, Vector researchers will participate in various workshops covering fairness in prediction, disentanglement in variational autoencoders, graph neural networks, and applications in cancer mutation trajectory reconstruction. Prominent contributors include James Martens, Jimmy Ba, Graham W. Taylor, and Richard S. Zemel. This participation underscores the institute's active role in advancing deep learning methodologies and their practical applications within the global artificial intelligence community.
Wire timeline
Vector Institute Researchers Present Latest AI Findings at ICLR 2018
The Vector Institute for Artificial Intelligence announced that its faculty and researchers will attend the International Conference on Learning Representations (ICLR) to present their latest advancements in machine learning. The institute highlighted a comprehensive list of accepted papers and workshop contributions scheduled for the conference in late April and early May 2018. Key research topics include Kronecker-factored curvature approximations for recurrent neural networks, quantitative evaluation of Generative Adversarial Networks (GANs), and efficient weight perturbations via Flipout. Other significant works address attacking binarized neural networks, meta-learning for few-shot classification, and optimizing control variates for black-box gradient estimation. Additionally, Vector researchers will participate in various workshops covering fairness in prediction, disentanglement in variational autoencoders, graph neural networks, and applications in cancer mutation trajectory reconstruction. Prominent contributors include James Martens, Jimmy Ba, Graham W. Taylor, and Richard S. Zemel. This participation underscores the institute's active role in advancing deep learning methodologies and their practical applications within the global artificial intelligence community.
Vector Institute for Artificial Intelligence