Skip to the content.

Overview

The workshop will take place on Friday May 7, 2021. Due to the pandemic, the workshop will be VIRTUAL. More details will be provided soon.

Note that to attend the event, a registration on the ICLR website is required. All workshop events (except Poster session and open discussion) can be followed using the ICLR link or use the zoom link by clicking on “join zoom” on the ICLR link. For the Poster session participants should une the Gather.town link. Note that papers id can be found on Accepter papers section and the workshop replay can be fond on the ICLR link.

Aim of the Workshop

To reach top-tier performance, deep learning architectures usually rely on a large number of parameters and operations, and thus require to be processed using considerable power and memory. Numerous works have proposed to tackle this problem using quantization of parameters, pruning, clustering of parameters, decompositions of convolutions or using distillation. However, most of these works aim at accelerating only the inference process and disregard the training phase. In practice, however, it is the learning phase that is by far the most complex. There has been recent efforts in introducing some compression on training process, however it remains challenging.

In this workshop, we propose to focus on reducing the complexity of the training process. Our aim is to gather researchers interested in reducing energy, time, or memory usage for faster/cheaper/greener prototyping or deployment of deep learning models. Due to the dependence of deep learning on large computational capacities, the outcomes of the workshop could benefit all who deploy these solutions, including those who are not hardware specialists. Moreover, it would contribute to making deep learning more accessible to small businesses and small laboratories.

Indeed, training complexity is of interest to many distinct communities. A first example is training on edge devices, where training can be used to specialize to data obtained online when the data cannot be transmitted back to the cloud because of constraints on privacy or communication bandwidth. Another example is accelerating training on dedicated hardware such as GPUs or TPUs.

Our Sponsors