---
title: "The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks"
description: "The Lottery Ticket Hypothesis asks why a small neural network can be difficult to train even when pruning proves that it has enough capacity. Read it to see how iterative magnitude pruning discovers a sparse mask, why resetting its surviving weights to their exact original values matters, and why the resulting ticket is still expensive to find and awkward for ordinary hardware."
canonical_url: "https://fanout.sh/daily/2026-09-13-lottery-ticket-hypothesis"
md_url: "https://fanout.sh/daily/2026-09-13-lottery-ticket-hypothesis.md"
last_updated: "2026-09-13"
access: "public"
---

# The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks

The Lottery Ticket Hypothesis asks why a small neural network can be difficult to train even when pruning proves that it has enough capacity. Read it to see how iterative magnitude pruning discovers a sparse mask, why resetting its surviving weights to their exact original values matters, and why the resulting ticket is still expensive to find and awkward for ordinary hardware.

## Paper details

- Authors: Jonathan Frankle and Michael Carbin

- Venue: ICLR 2019

- Track: AI Research

- Difficulty: Easy

- Reading time: 30 min

- Original paper: https://arxiv.org/abs/1803.03635v5

## What you will learn

- Why a winning ticket is the combination of a sparse mask and the surviving weights' original initialization, not the sparse topology alone.

- How train, prune, reset, and repeat turns post-training magnitude into a retrospective search signal for a trainable subnetwork.

- What the MNIST and CIFAR-10 results show, and how repeated training, unstructured sparsity, learning-rate sensitivity, and untested generality limit the claim.

## Continue exploring

[Explore neural-network pruning in Fanout](https://fanout.sh/knowledge-graph?node=ai-research%3Aconcept%3Apruning): Connect the paper's binary masks and sparse subnetworks to Fanout's pruning concept, model compression, distillation, quantization, and the difference between fewer parameters and faster execution.

---
This representation contains public Fanout content only. Protected Pro lessons, account data, billing, checkout, and pricing are not included.

Browse the public content map: https://fanout.sh/sitemap.md