Biathlon: Harnessing Model Resilience for Accelerating ML Inference Pipelines

May 18, 2024 Β· Declared Dead Β· πŸ› Proceedings of the VLDB Endowment

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Chaokun Chang, Eric Lo, Chunxiao Ye arXiv ID 2405.11191 Category cs.DB: Databases Cross-listed cs.LG Citations 3 Venue Proceedings of the VLDB Endowment Last Checked 5 months ago
Abstract
Machine learning inference pipelines commonly encountered in data science and industries often require real-time responsiveness due to their user-facing nature. However, meeting this requirement becomes particularly challenging when certain input features require aggregating a large volume of data online. Recent literature on interpretable machine learning reveals that most machine learning models exhibit a notable degree of resilience to variations in input. This suggests that machine learning models can effectively accommodate approximate input features with minimal discernible impact on accuracy. In this paper, we introduce Biathlon, a novel ML serving system that leverages the inherent resilience of models and determines the optimal degree of approximation for each aggregation feature. This approach enables maximum speedup while ensuring a guaranteed bound on accuracy loss. We evaluate Biathlon on real pipelines from both industry applications and data science competitions, demonstrating its ability to meet real-time latency requirements by achieving 5.3x to 16.6x speedup with almost no accuracy loss.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Databases

Died the same way β€” πŸ‘» Ghosted