Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning
Learning Deep Parsimonious Representations
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 |
|---|---|---|---|
| add sample-clustering regularization to autoencoder | autoencoder · MNIST test | reconstruction error: 3.61 -> 3.50 | Helped |
| add sample-clustering regularization to autoencoder | autoencoder · MNIST train | reconstruction error: 2.69 -> 2.73 | Hurt |
| add sample-clustering regularization to autoencoder | autoencoder · MNIST | std of error: 0.13 -> 0.01 | Helped |
| add sample-clustering to Caffe quick CIFAR10 net | Caffe CIFAR10 · CIFAR10 test | accuracy: 76.32 -> 81.05 | Helped |
| add sample-clustering to Caffe quick CIFAR100 net | Caffe CIFAR100 · CIFAR100 test | accuracy: 46.21 -> 50.50 | Helped |
| clustering regularizer versus Dropout | Caffe CIFAR10 · CIFAR10 test | accuracy: 77.45 -> 81.05 | Helped |
| clustering regularizer versus DeCov | Caffe CIFAR100 · CIFAR100 test | accuracy: 40.34 -> 50.50 | Helped |
| swap conv clustering objective sample vs spatial vs channel co-clustering (ablation) | Caffe CIFAR10 · CIFAR10 test | accuracy: 81.05 -> 80.65 | No clear effect |
| swap conv clustering objective on CIFAR100 (ablation) | Caffe CIFAR100 · CIFAR100 test | accuracy: 50.50 -> 49.80 | No clear effect |
| add sample clustering, measure K-means NMI on fc4 features | Caffe CIFAR100 · CIFAR100 | NMI: 0.4122 -> 0.4914 | Helped |
| add clustering to fine-tuned AlexNet versus DeCAF | AlexNet · CUB-200-2011 | accuracy: 58.75 -> 61.77 | Helped |
| sample-clustering features versus SJE in zero-shot | AlexNet fc7 · CUB zero-shot | top-1 accuracy: 40.3 -> 46.1 | Helped |
Closest papers by meaning
- Unsupervised Deep Embedding for Clustering Analysis (ICML 2016 · similarity 0.86)
- Towards K-means-friendly Spaces: Simultaneous Deep Learning and Clustering (ICML 2017 · similarity 0.86)
- Intraclass clustering: an implicit learning ability that regularizes DNNs (ICLR 2021 · similarity 0.85)
- Exploring a Principled Framework for Deep Subspace Clustering (ICLR 2025 · similarity 0.84)
- Unveiling the Unseen: Identifiable Clusters in Trained Depthwise Convolutional Kernels (ICLR 2024 · similarity 0.84)
- Hard Regularization to Prevent Deep Online Clustering Collapse without Data Augmentation (AAAI 2024 · similarity 0.83)
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.