Deep Learning
Semi-supervised Deep Kernel Learning: Regression with Unlabeled Data by Minimizing Predictive Variance
Neal Jean, Sang Michael Xie, Stefano Ermon
Large amounts of labeled data are typically required to train deep learning models.
Pelee: A Real-Time Object Detection System on Mobile Devices
Jun Wang, Tanner Bohn, Charles Ling
However, all these models are heavily dependent on depthwise separable convolution which lacks efficient implementation in most deep learning frameworks.
Generalisation of structural knowledge in the hippocampal-entorhinal system
James Whittington, Timothy Muller, Shirely Mark, Caswell Barry, Tim Behrens
A central problem to understanding intelligence is the concept of generalisation. This allows previously learnt structure to be exploited to solve tasks in novel situations differing in their particularities.
L4: Practical loss-based stepsize adaptation for deep learning
We propose a stepsize adaptation scheme for stochastic gradient descent.
Learning semantic similarity in a continuous space
Neural Information Processing Systems http://nips.cc/
Turbo Learning for CaptionBot and DrawingBot
Qiuyuan Huang, Pengchuan Zhang, Dapeng Wu, Lei Zhang
Neural Information Processing Systems http://nips.cc/
Constructing Fast Network through Deconstruction of Convolution
By using deeper and wider networks, network accuracy can be improved rapidly. However, in an environment with limited resources (e.g., mobile applications), heavy networks may not be usable.
Training Deep Models Faster with Robust, Approximate Importance Sampling
Tyler B. Johnson, Carlos Guestrin
Neural Information Processing Systems http://nips.cc/
Constructing Deep Neural Networks by Bayesian Network Structure Learning
Raanan Y. Rohekar, Shami Nisimov, Yaniv Gurwicz, Guy Koren, Gal Novik
We introduce a principled approach for unsupervised structure learning of deep neural networks.