CVPR 2019 Attracts 9K Attendees; Best Papers Announced; ImageNet Honoured 10 Years Later
Conference organizers have announced the recipient of the CVPR 2019 Best Paper Award: A Theory of Fermat Paths for Non-Line-of-Sight Shape Reconstruction from Carnegie Mellon University, University of Toronto, and University College London. The paper presents a novel theory on Fermat paths of light between a known visible scene and an unknown object not in the line of sight of a transient camera.
CVPR 2019 | Synced Notable Paper Picks
– Envisioning Privacy Preserving Image-Based Localization for Augmented Reality
– A Theory of Fermat Paths for Non-Line-of-Sight Shape Reconstruction
– Neural Task Graphs: Generalizing to Unseen Tasks from A Single Video Demonstration
– Using AI to Generate Recipes from Food Images
– MediaPipe: A Framework for Perceiving and Augmenting Reality
– Using Platform-Aware AI to Design Compact and Efficient Neural Networks
– Accelerating MRI Reconstruction via Active Acqui
Functional Regularisation for Continual Learning
Researchers introduce a framework for continual learning based on Bayesian inference over the function space rather than the parameters of a deep neural network. This method, referred to as functional regularisation for continual learning, avoids forgetting a previous task by constructing and memorising an approximate posterior belief over the underlying task-specific function.
Meta-Learning Surrogate Models for Sequential Decision Making
Researchers introduce a unified probabilistic framework for solving sequential decision making problems ranging from Bayesian optimisation to contextual bandits and reinforcement learning. This is accomplished by a probabilistic model-based approach that explains observed data while capturing predictive uncertainty during the decision making process.
Bridging the Domain Gap for Neural Models
To understand the challenge behind domain shift and the need for domain adaptation, researchers establish a simple pilot experiment: they use the real-world house number images from SVHN dataset as one domain and the handwritten digit images from MNIST dataset as another domain.
You May Also Like
If Tom Brings Jerry Home, ML Locks the Cat Door
Amazon Sr. Product Manager Benjamin Hamm had a problem with his cat “Metric,” a keen hunter who brings a catch into the house roughly every 10 days. Hamm addressed the problem by learning to code, and shared the ML solution he developed in a recent and amusing Ignite Seattle talk, “Cats, Rats, A.I., Oh My!”
Leading Researchers Publish ‘Climate Change + AI’ Document
In an unprecedented “call for collaboration,” a group of 22 respected AI experts that includes Andrew Ng, Yoshua Bengio, and Demis Hassabis have published a paper exploring how machine learning (ML) could help deal with climate change by reducing greenhouse gases (GHG) and proposing how societies might initiate and adapt to these changes.
Global AI Events
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
September 10-12: The AI Summit (Part of TechXLR8) in Singapore
September 24-28: Microsoft Ignite in Orlando, United States
Global AI Opportunities