MLconf NYC 2015
| title | author | length |
|---|---|---|
| Jeremy Schiff, Senior Manager, Data Science, OpenTable | Jeremy Schiff | 20m |
OpenTable’s recommender combines an IR search stack with collaborative filtering, serializes CF models for real‑time scoring and uses Spark for batch tuning… | ||
| Claudia Perlich, Chief Scientist, Dstillery | Claudia Perlich | 13m |
Argues that optimizing for clicks backfires because click is a terrible metric — optimization surfaces non‑purchase behaviors (flashlight apps, motor issues)… | ||
| Jeremy Stanley, EVP/Data Scientist, Sailthru | Jeremy Stanley | 24m |
Sailthru launches Sightlines to predict per-user outcomes across ~5M users per client and ~1,000 clients, building ~10 models per client daily; the pipeline… | ||
| Juliet Hougland, Data Scientist, Cloudera | Juliet Hougland | 22m |
Matrix Decomposition at Scale: Matrix decomposition is an incredibly common task in machine learning, appearing everywhere including recommendation algorithms… | ||
| Ronald Menich, Chief Data Scientist, Predictix, LLC | Ronald Menich | 22m |
Claims that replacing legacy time‑series stacks with factorization machines plus heavy feature engineering yields more accurate, extensible retail forecasts… | ||
| Corinna Cortes, Head of Research, Google | Corinna Cortes | 37m |
Google deploys Structured Snippets (launched Q3 2014) that extract facts from web tables and text, and combines them into probabilistic facts via a Knowledge… | ||
| Jeff Johnson, Research Engineer, Facebook | Jeff Johnson | 19m |
Argues that many deep learning layers are too small for GPUs, shows an open-source Torch convolution engine (FB FFT) that reorders work to fit registers and… | ||
| Ted Willke, Senior Principal Engineer, Intel Labs | Ted Willke | 26m |
Claims real-time fMRI decoding can distinguish covert attention to faces versus scenes with around 95% accuracy by running MVPA and SVM on voxel time series… | ||
| nothing matches | ||