AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Snap ML: A Hierarchical Framework for Machine Learning

Celestine Dünner, Thomas Parnell, Dimitrios Sarigiannis, Nikolas Ioannou, Andreea Anghel, Gummadi Ravi, Madhusudanan Kandasamy, Haralampos Pozidis

NeurIPS 2018 · Learning theory and algorithms › Prediction and decision losses · approach: Active learning and subset selection · 8 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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 changedWhereMeasuredEffect
Use Snap ML instead of scikit-learnLogistic regression training · criteo-kaggletraining time: around 200 seconds -> around 20 secondsImproved
Use Snap ML instead of TensorFlowLogistic regression training · criteo-kaggletraining time: approximately 500 seconds -> around 20 secondsImproved
Use NVLINK 2.0 for pipelined data streaming instead of the PCI-based setupSnap ML streaming · criteo-tb first 200 million training examplesspeedup: 3.5xImproved
Increase inner CoCoA iterations t_2 above 1Hierarchical CoCoA · criteo-tb first billion examples, InfiniBandtime to target training suboptimalityNo clear effect
Increase inner CoCoA iterations t_2 above 1Hierarchical CoCoA · criteo-tb first billion examples, 1Gbit Ethernettime to target training suboptimalityImproved
Use Snap ML instead of the best previously reported TensorFlow resultLogistic regression · criteo-tbtraining speed: 46x fasterImproved
Use Snap ML instead of TensorFlow on the same hardwareLogistic regression · criteo-tbtraining speed: over 500× fasterImproved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy