Timezone: »

Rare Gems: Finding Lottery Tickets at Initialization
Kartik Sreenivasan · Jy-yong Sohn · Liu Yang · Matthew Grinde · Alliot Nagle · Hongyi Wang · Eric Xing · Kangwook Lee · Dimitris Papailiopoulos

Tue Nov 29 02:00 PM -- 04:00 PM (PST) @ Hall J #639
Large neural networks can be pruned to a small fraction of their original size, with little loss in accuracy, by following a time-consuming "train, prune, re-train" approach. Frankle & Carbin conjecture that we can avoid this by training lottery tickets, i.e., special sparse subnetworks found at initialization, that can be trained to high accuracy. However, a subsequent line of work presents concrete evidence that current algorithms for finding trainable networks at initialization, fail simple baseline comparisons, e.g., against training random sparse subnetworks. Finding lottery tickets that train to better accuracy compared to simple baselines remains an open problem. In this work, we resolve this open problem by proposing Gem-Miner which finds lottery tickets at initialization that beat current baselines. Gem-Miner finds lottery tickets trainable to accuracy competitive or better than Iterative Magnitude Pruning (IMP), and does so up to $19\times$ faster.

Author Information

Kartik Sreenivasan (University of Wisconsin-Madison)
Jy-yong Sohn (University of Wisconsin-Madison)
Liu Yang (University of Wisconsin, Madison)
Matthew Grinde (University of Wisconsin - Madison)
Alliot Nagle (UT Austin)

ECE PhD @ UT Austin

Hongyi Wang (Carnegie Mellon University)
Eric Xing (Petuum Inc.)
Kangwook Lee (UW Madison, Krafton)
Dimitris Papailiopoulos (University of Wisconsin-Madison)

More from the Same Authors