Neural Information Processing Systems http://nips.cc/

In the real-word applications, data is usually represented in different views, including multiple modalities or multiple types of features.

Ultimately, machine learning models are used to make decisions.

The recent "Lottery Ticket Hypothesis" paper by Frankle & Carbin showed that a

Neural Processes combine the strengths of neural networks and Gaussian processes to achieve both flexible learning and fast prediction in stochastic processes.

This paper introduces in-place zero-space ECC assisted with a new training scheme weight distribution-oriented training .

In our experiments, we show that hyperbolic GNNs can lead to substantial improvements on various benchmark datasets.

Previous literature offers limited clues on how to learn a periodic function using modern neural networks.

We introduce Natural Neural Networks, a novel family of algorithms that speed up convergence by adapting their internal representation during training to improve conditioning of the Fisher matrix. In particular, we show a specific example that employs a simple and efficient reparametrization of the neural network weights by implicitly whitening the representation obtained at each layer, while preserving the feed-forward computation of the network. Such networks can be trained efficiently via the proposed Projected Natural Gradient Descent algorithm (PRONG), which amortizes the cost of these reparametrizations over many parameter updates and is closely related to the Mirror Descent online learning algorithm. We highlight the benefits of our method on both unsupervised and supervised learning tasks, and showcase its scalability by training on the large-scale ImageNet Challenge dataset.