Difference between revisions of "LightR"
Jump to navigation
Jump to search
m |
m |
||
| Line 15: | Line 15: | ||
: Hand-crafted features with naive KNN -> Search-based sequence model | : Hand-crafted features with naive KNN -> Search-based sequence model | ||
: Slow networks -> Knowedge distill & Quantization aware training | : Slow networks -> Knowedge distill & Quantization aware training | ||
| + | : Offline pretraining & online fine-tuning of everything above. | ||
Revision as of 08:15, 11 February 2022
- LightR Sub-pages:
- Version History
This page is under construction. For recent activities, see Version History.
- Design principle
- Strategy light, machine learning heavy.
- Planned experiments
- Multiple hand-tuned surfing models -> Expert network & gate network
- Hand-crafted features with naive KNN -> Search-based sequence model
- Slow networks -> Knowedge distill & Quantization aware training
- Offline pretraining & online fine-tuning of everything above.
