Research map / Architectures and efficiency / Quantization
Flexpoint: An Adaptive Numerical Format for Efficient Training of Deep Neural Networks
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 |
|---|---|---|---|
| train with Flexpoint flex16+5 instead of float32 | AlexNet · ImageNet1k | top-5 misclassification | No clear effect |
| train with Flexpoint flex16+5 instead of float32 | 110-layer ResNet · CIFAR-10 validation | top-1 misclassification | No clear effect |
| train with float16 instead of float32, without changing hyperparameters | AlexNet and 110-layer ResNet · ImageNet1k and CIFAR-10 validation | misclassification | Hurt |
| train with Flexpoint flex16+5 instead of float32 | Wasserstein GAN · LSUN bedroom | Wasserstein distance | No clear effect |
| train with Flexpoint flex16+5 instead of float32 | Wasserstein GAN · LSUN bedroom | Fréchet Inception Distance | No clear effect |
| train with float16 instead of float32, without changing hyperparameters | Wasserstein GAN · LSUN bedroom | Wasserstein distance | Hurt |
| train with float16 instead of float32, without changing hyperparameters | Wasserstein GAN · LSUN bedroom | Fréchet Inception Distance | Hurt |
Closest papers by meaning
- Be Like Water: Adaptive Floating Point for Machine Learning (ICML 2022 · similarity 0.87)
- FlexRound: Learnable Rounding based on Element-wise Division for Post-Training Quantization (ICML 2023 · similarity 0.86)
- Stepping Forward on the Last Mile (NeurIPS 2024 · similarity 0.85)
- A Block Minifloat Representation for Training Deep Neural Networks (ICLR 2021 · similarity 0.85)
- Tensor Programs VI: Feature Learning in Infinite Depth Neural Networks (ICLR 2024 · similarity 0.84)
- Random Sparse Lifts: Construction, Analysis and Convergence of finite sparse networks (ICLR 2024 · similarity 0.84)
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.