Research map / Architectures and efficiency / Pruning and efficient tuning
Exploring Lottery Ticket Hypothesis in Spiking Neural Networks
Youngeun Kim, Yuhang Li, Hyoungseob Park, Yeshwanth Venkatesha, Ruokai Yin, Priyadarshini Panda
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add Early-Time search to IMP (ablation) | IMP · CIFAR10 VGG16 (98.13% sparsity) | accuracy and search speed: -0.71 acc, x1.35 speed | Mixed |
| add Early-Time search to IMP (ablation) | IMP · CIFAR10 ResNet19 (98.13% sparsity) | accuracy and search speed: -1.07 acc, x1.58 speed | Mixed |
| add Early-Time search to IMP (ablation) | IMP · CIFAR100 ResNet19 (98.13% sparsity) | accuracy and search speed: -1.86 acc, x1.57 speed | Mixed |
| add Early-Time search to IMP (ablation) | IMP · CIFAR100 VGG16 (98.13% sparsity) | accuracy and search speed: -1.61 acc, x1.36 speed | Mixed |
| add Early-Time search to Early-Bird ticket (ablation) | EB ticket · CIFAR10 VGG16 (98.13% sparsity) | accuracy and search speed: +0.22 acc, x1.12 speed | Mixed |
| add Early-Time search to Early-Bird ticket (ablation) | EB ticket · CIFAR100 ResNet19 (98.13% sparsity) | accuracy and search speed: -1.42 acc, x1.16 speed | Mixed |
| use Early-Bird ticket instead of IMP | IMP · CIFAR10 VGG16 (98.13% sparsity) | accuracy (%): 91.81 -> 84.64 | Hurt |
| use global instead of local pruning (ablation) | IMP · CIFAR10 VGG16 | accuracy (%) | Improved |
| use late rewinding in IMP (ablation) | IMP · deep SNNs | accuracy (%) | Improved |
| rewind to epoch 260 instead of early epoch (ablation) | IMP · CIFAR10 VGG16 (98.13% sparsity) | accuracy (%): ~1.5% drop at epoch 260 | Hurt |
| transfer ANN winning ticket to SNN | IMP ticket from ANN · SVHN and Fashion-MNIST (98.13% sparsity) | accuracy (%): less than 2% drop at 98.13% sparsity | Improved |
| transfer ANN winning ticket to SNN | IMP ticket from ANN · CIFAR10 and CIFAR100 | accuracy (%) | Hurt |
Closest papers by meaning
- The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks (ICLR 2019 · similarity 0.91)
- Why Lottery Ticket Wins? A Theoretical Perspective of Sample Complexity on Sparse Neural Networks (NeurIPS 2021 · similarity 0.90)
- Winning Lottery Tickets in Deep Generative Models (AAAI 2021 · similarity 0.90)
- Dual Lottery Ticket Hypothesis (ICLR 2022 · similarity 0.90)
- Plant 'n' Seek: Can You Find the Winning Ticket? (ICLR 2022 · similarity 0.90)
- Find A Winning Sign: Sign Is All We Need to Win the Lottery (ICLR 2025 · similarity 0.88)
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.
