Research map / Learning theory and algorithms / Prediction and decision losses
Locality Sensitive Teaching
Zhaozhuo Xu, Beidi Chen, Chaojian Li, Weiyang Liu, Le Song, Yingyan Lin, Anshumali Shrivastava
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 |
|---|---|---|---|
| replace IMT exact greedy selection with LSH-based sampling (LST) | IMT · slice | seconds per iteration: 404.95 -> 0.196 | Improved |
| replace IMT with LST | IMT · 4 regression datasets | time to same train loss: 2000x faster | Improved |
| replace IMT with LST | IMT · abalone on TX2 | speedup: 425.12× | Improved |
| replace IMT with LST | IMT · abalone on TX2 | energy savings: 99.76% | Improved |
| replace IMT with LST | IMT · space ga on TX2 | speedup: 149.07× | Improved |
| replace IMT with LST | IMT · mg on TX2 | speedup: 117.20× | Improved |
| replace IMT with LST | IMT · synthetic classification on TX2 | speedup: 306.12× | Improved |
| replace IMT with LST | IMT · regression datasets | iteration-wise train and test loss | No clear effect |
| replace stochastic SGD teaching with LST or IMT | learner · regression datasets | iterations to converge | Improved |
| replace IMT with LST on slice | IMT · slice | iteration-wise loss: about 30% LST instances outperform IMT | Mixed |
| replace GTA with AIPS (randomized sampling) in toy example (ablation) | GTA · toy logistic regression | train loss per iteration | Improved |
| run LST on TX2 instead of server | LST · regression datasets | iteration-wise convergence | No clear effect |
Closest papers by meaning
- Nonparametric Iterative Machine Teaching (ICML 2023 · similarity 0.81)
- Foundations of Autonomous Vehicles: A Curriculum Model for Developing Competencies in Artificial Intelligence and the Internet of Things for Grades 7–10 (AAAI 2024 · similarity 0.79)
- Smart Motor: A Low-Cost Hardware and Software Toolkit for Introducing Supervised Machine Learning to Elementary School Students (AAAI 2025 · similarity 0.79)
- Personalized Learning in Action: Exploring AI and Robotics for Early Childhood Education (AAAI 2025 · similarity 0.76)
- TinyTrain: Resource-Aware Task-Adaptive Sparse Training of DNNs at the Data-Scarce Edge (ICML 2024 · similarity 0.76)
- FedSPU: Personalized Federated Learning for Resource-Constrained Devices with Stochastic Parameter Update (AAAI 2025 · similarity 0.76)
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.
