Research map / Label-efficient and robust learning / Test-time adaptation
Cockpit: A Practical Debugging Tool for the 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 |
|---|---|---|---|
| Track instruments jointly (business configuration) instead of summing individual instruments | Cockpit · CIFAR-10 3c3d | training-time overhead: 200\,\% -> 140\,\% | Helped |
| Parallelize the ten extra backward passes for Hessian quantities | Cockpit · CIFAR-10 3c3d | training-time overhead: less than 100\,\% | Helped |
| Track the full configuration every 64th iteration | Cockpit full view · CIFAR-10 3c3d | training time: every $64$-th | No clear effect |
| Track any configuration once per epoch | Cockpit · CIFAR-10 3c3d | training-time overhead | No clear effect |
Closest papers by meaning
- PyTorch: An Imperative Style, High-Performance Deep Learning Library (NeurIPS 2019 · similarity 0.84)
- SPADE: Sparsity-Guided Debugging for Deep Neural Networks (ICML 2024 · similarity 0.83)
- Maestro: Uncovering Low-Rank Structures via Trainable Decomposition (ICML 2024 · similarity 0.83)
- HiBug2: Efficient and Interpretable Error Slice Discovery for Comprehensive Model Debugging (ICLR 2025 · similarity 0.82)
- PAGER: Accurate Failure Characterization in Deep Regression Models (ICML 2024 · similarity 0.81)
- Unveiling AI's Blind Spots: An Oracle for In-Domain, Out-of-Domain, and Adversarial Errors (ICML 2025 · similarity 0.81)
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.