2022/06/16 by Surya Kant Sahu, Sahu, Surya Kant, Sai Mitheran +3
Computer Science · Engineering · Mathematics · #Artificial intelligence #Artificial neural network #Computer network #Computer science #Data mining #Econometrics #Engineering #FOS: Computer and information sciences #Initialization #Lottery #Machine Learning (cs.LG) #Machine Learning and Algorithms #Machine Learning and Data Classification #Mathematics #Neural Networks and Applications #Relation (database) #Statistics #Ticket #Work (physics) #cs.LG
paper · pdf · doi:10.48550/arxiv.2206.08175
published in arXiv (Cornell University) (Cornell University) · Accepted at ICML 2022 HAET Workshop
arxiv created 2022/06/16 · openalex publication_date 2022/06/16 · arxiv updated 2022/06/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
The Lottery Ticket Hypothesis (LTH) states that for a reasonably sized neural network, a sub-network within the same network yields no less performance than the dense counterpart when trained from the same initialization. This work investigates the relation between model size and the ease of finding these sparse sub-networks. We show through experiments that, surprisingly, under a finite budget, smaller models benefit more from Ticket Search (TS).