497 citations
- Georgia Institute of TechnologyUS50 papers
- Microsoft (United States)US4 papers
- Carnegie Mellon UniversityUS3 papers
- Emory UniversityUS3 papers
- Massachusetts Institute of TechnologyUS3 papers
- Microsoft Research (United Kingdom)GB3 papers
- University of California, BerkeleyUS3 papers
- University of California San DiegoUS3 papers
- University of Wisconsin–MadisonUS3 papers
- Google (United States)US2 papers
- IBM (United States)US2 papers
- Princeton UniversityUS2 papers
5 papers · 1 filter
A demonstrator for a real-time AI-FPGA-based triggering system for sPHENIX at RHIC
J. Kvapil, G. Borca-Tasciuc, H. Bossi +28
The RHIC interaction rate at sPHENIX will reach around 3 MHz in pp collisions and requires the detector readout to reject events by a factor of over 200 to fit the DAQ bandwidth of…
Apparate: Rethinking Early Exits to Tame Latency-Throughput Tensions in ML Serving
Yinwei Dai, Rui Pan, Anand Iyer +2
Machine learning (ML) inference platforms are tasked with balancing two competing goals: ensuring high throughput given many requests, and delivering low-latency responses to suppo…
How to Estimate Model Transferability of Pre-Trained Speech Models?
Zih-Ching Chen, Chao-Han Huck Yang, Bo Li +6
In this work, we introduce a "score-based assessment" framework for estimating the transferability of pre-trained speech models (PSMs) for fine-tuning target tasks. We leverage upo…
Angler: Helping Machine Translation Practitioners Prioritize Model Improvements
Samantha Robertson, Zijie J. Wang, Dominik Moritz +2
Machine learning (ML) models can fail in unexpected ways in the real world, but not all model failures are equal. With finite time and resources, ML practitioners are forced to pri…
GAM Coach: Towards Interactive and User-centered Algorithmic Recourse
Zijie J. Wang, Jennifer Wortman Vaughan, Rich Caruana +1
Machine learning (ML) recourse techniques are increasingly used in high-stakes domains, providing end users with actions to alter ML predictions, but they assume ML developers unde…