Subscribe to Synced Global AI Weekly
ICML 2019 | Google, ETH Zurich, MPI-IS, Cambridge & PROWLER.io Share Best Paper Honours
ICML announced the recipients of the Best Paper Awards: Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations from Google Research, ETH Zurich, and Max Planck Institute for Intelligent Systems; and Rates of Convergence for Sparse Variational Gaussian Process Regressionfrom the University of Cambridge and PROWLER.io.
(Synced)
ICML 2019 | Good Papers Collection
– Google at ICML 2019
– Facebook Research at ICML 2019
– Intel AI Research at ICML 2019
– Google Brain | Similarity of Neural Network Representations Revisited
– Georgia Institute of Technology & Ant Financial |Generative Adversarial User Model for Reinforcement Learning Based Recommendation System
– CMU & UC Berkeley | Learning Correspondence from the Cycle-consistency of Time
– BAIR | Learning to Learn with Probabilistic Task Embeddings
What to Expect at CVPR 2019
– Microsoft at CVPR 2019
– IBM Research AI at CVPR 2019
– Baidu at CVPR 2019
– NVIDIA Research at CVPR 2019
– Intel AI Research at CVPR 2019
– Facebook AI | Creating 2.5D Visual Sound for An Immersive Audio Experience
– 2019 CVPR Accepted Papers Organized in A Parsable and Easier to Sort Through Way
Technology
What Does BERT Look At? An Analysis of BERT’s Attention
Researchers propose methods for analyzing the attention mechanisms of pre-trained models and apply them to BERT. BERT’s attention heads exhibit patterns such as attending to delimiter tokens, specific positional offsets, or broadly attending over the whole sentence, with heads in the same layer often exhibiting similar behaviors.
(Stanford University & Facebook AI Research)
Are Weights Really Important to Neural Networks?
As with the age-old “nature versus nurture” debate, AI researchers want to know whether architecture or weights play the main role in the performance of neural networks. In a blow to the “nurture” side, Google researchers have now demonstrated that a neural network which has not learned weights through training can still achieve satisfactory results in machine learning tasks.
(Synced) /(Google Brain)
Episodic Memory in Lifelong Language Learning
Researchers introduce a lifelong language learning setup where a model needs to learn from a stream of text examples without any dataset identifier. They propose an episodic memory model that performs sparse experience replay and local adaptation to mitigate catastrophic forgetting in this setup.
(DeepMind)
You May Also Like
New Facebook PyTorch Hub Facilitates Reproducibility Testing
In a bid to provide a smoother reproduction experience, Facebook has announced the beta release of PyTorch Hub, a new pretrained model repository designed to facilitate research reproducibility testing.
(Synced)
BatchNorm + Dropout = DNN Success!
A group of researchers from Tencent Technology, the Chinese University of Hong Kong, and Nankai University recently combined two commonly used techniques — Batch Normalization (BatchNorm) and Dropout — into an Independent Component (IC) layer inserted before each weight layer to make inputs more independent.
(Synced)
Global AI Events
June 15-21: Computer Vision and Pattern Recognition in Long Beach, United States
June 20-21: AI for Good Summit in San Francisco, United States
June 28: Research and Applied AI Summit (RAAIS) in London, United Kingdom
August 19-23: Knowledge Discovery and Data Mining (KDD2019) in London, United Kingdom
Global AI Opportunities
Research Scientist, Google Brain Toronto
OpenAI Seeking Software Engineers and Deep Learning Researchers
DeepMind Scholarship: Access to Science
Postdoctoral Researcher (AI) – Self-Supervised Learning
Pingback: ICML Announces Best Paper Awards; What to Expect at CVPR 2019 - Connected Free