Research map / Architectures and efficiency / RNNs and normalization
Adaptive Classification for Prediction Under a Budget
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 adaptive gating and low-cost prediction to bypass the high-cost model on suitable inputs | Adapt-GBRT · MiniBooNE | average feature cost: 63% | Helped |
| Use adaptive gating and low-cost prediction to bypass the high-cost model on suitable inputs | Adapt-GBRT · Forest | average feature cost: 32% | Helped |
| Use adaptive gating and low-cost prediction to bypass the high-cost model on suitable inputs | Adapt-GBRT · Yahoo! | average feature cost: 58% | Helped |
| Use adaptive gating and low-cost prediction to bypass the high-cost model on suitable inputs | Adapt-GBRT · CIFAR-10 | average feature cost: 12% | Helped |
| Use adaptive gating and low-cost prediction to bypass the high-cost model on suitable inputs | Adapt-GBRT · Letters | average feature cost: 31% | Helped |
| Jointly optimize feature selection for the gate and low-cost predictor rather than selecting a sparse feature set first | Adapt-Lin · Synthetic-2 | average feature cost: 1.75 -> 1.5 | Helped |
| Use a bottom-up adaptive approximation of the high-accuracy model rather than a stand-alone top-down budgeted model | Adapt-GBRT · MiniBooNE, Forest, Yahoo!, and CIFAR-10 | accuracy-cost tradeoff | Helped |
| Use a nonlinear boosted-tree gate and predictor instead of linear classifiers | Adapt-GBRT · Letters | accuracy-cost tradeoff | Helped |
Closest papers by meaning
- Protected Test-Time Adaptation via Online Entropy Matching: A Betting Approach (NeurIPS 2024 · similarity 0.83)
- TinyTTA: Efficient Test-time Adaptation via Early-exit Ensembles on Edge Devices (NeurIPS 2024 · similarity 0.82)
- TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge (ICML 2024 · similarity 0.82)
- Learning with Adaptive Resource Allocation (ICML 2024 · similarity 0.81)
- Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment (NeurIPS 2025 · similarity 0.81)
- A Layer Selection Approach to Test Time Adaptation (AAAI 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.