ICLR 2020

Apple sponsored the 8th International Conference on Learning Representations (ICLR) in April 2020, which took place virtually from April 26 - May 1. ICLR focuses on the advancement of representation learning, and this year’s conference included presentations on cutting-edge research on deep learning areas including computer vision, text understanding, data science, and more.

Learn more about ICLR

See our teams at Jobs at Apple

Accepted Papers

Stochastic Weight Averaging in Parallel: Large-Batch Learning That Generalizes Well

Vipul Gupta, Santiago Akle Serrano, Dennis DeCoste

We propose Stochastic Weight Averaging in Parallel (SWAP), an algorithm to ac- celerate DNN training. Our algorithm uses large mini-batches to compute an approximate solution quickly and then refines it by averaging the weights of multiple models computed independently and in parallel. The resulting models generalize equally well as those trained with small mini-batches but are produced in a sub- stantially shorter time. We demonstrate the reduction in training time and the good generalization performance of the resulting models on the computer vision datasets CIFAR10, CIFAR100, and ImageNet.

Capsules with Inverted Dot-Product Attention Routing

Yao-Hung Hubert Tsai, Nitish Srivastava, Hanlin Goh, Ruslan Salakhutdinov

We introduce a new routing algorithm for capsule networks, in which a child capsule is routed to a parent based only on agreement between the parent’s state and the child’s vote. The new mechanism 1) designs routing via inverted dot-product attention; 2) imposes Layer Normalization as normalization; and 3) replaces sequential iterative routing with concurrent iterative routing. When compared to previously proposed routing algorithms, our method improves performance on benchmark datasets such as CIFAR-10 and CIFAR-100, and it performs at-par with a powerful CNN (ResNet-18) with 4× fewer parameters. On a different task of recognizing digits from over-layed digit images, the proposed capsule model performs favorably against CNNs given the same number of layers and neurons per layer. We believe that our work raises the possibility of applying capsule networks to complex real-world tasks.