Research map / Architectures and efficiency / Pruning and efficient tuning
Plant 'n' Seek: Can You Find the Winning Ticket?
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use singleshot pruning instead of multishot pruning | GraSP, SNIP, Synflow, Magnitude, and Random · Circle and ReLU | ticket performance | Mixed |
| use singleshot pruning to find weak tickets | GraSP, SNIP, Synflow, Magnitude, and Random · Helix | ticket performance | Hurt |
| increase singleshot pruning sparsity to the extreme low-sparsity range | GraSP, SNIP, Synflow, Magnitude, and Random · Circle, ReLU, and Helix | ticket performance | Hurt |
| add multishot training, pruning, and resetting | Magnitude, SNIP, and Synflow · Circle | weak-ticket performance | Helped |
| add multishot training, pruning, and resetting | Magnitude, SNIP, and Synflow · Helix | weak-ticket performance | Helped |
| use multishot pruning | Synflow · Circle | weak-ticket performance | Helped |
| use multishot pruning | GraSP · Circle, ReLU, and Helix | ticket performance | Hurt |
| train edge-popup for 10 rounds | edge-popup · Circle | strong-ticket performance | Mixed |
| anneal the target sparsity across rounds (ablation) | edge-popup · Circle | strong-ticket performance | Helped |
| apply edge-popup to regression tasks | edge-popup · ReLU and Helix | ticket performance | Hurt |
| add layer-wise local sparsity constraints (ablation) | GraSP with multishot pruning · Circle and Helix | ticket performance | No clear effect |
| remove noise from the data (ablation) | singleshot pruning methods · Circle | ticket performance | No clear effect |
Closest papers by meaning
- The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks (ICLR 2019 · similarity 0.94)
- Winning Lottery Tickets in Deep Generative Models (AAAI 2021 · similarity 0.93)
- Winning the Lottery with Continuous Sparsification (NeurIPS 2020 · similarity 0.93)
- Why Lottery Ticket Wins? A Theoretical Perspective of Sample Complexity on Sparse Neural Networks (NeurIPS 2021 · similarity 0.93)
- Find A Winning Sign: Sign Is All We Need to Win the Lottery (ICLR 2025 · similarity 0.90)
- Finding Lottery Tickets in Vision Models via Data-driven Spectral Foresight Pruning (CVPR 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.