MLconf NYC 2017
| title | author | length |
|---|---|---|
| Irina Rish, Researcher, IBM Watson, NYC 2017 | Irina Rish | 34m |
Applies machine learning to neuroimaging with concrete results: fMRI functional-network features classify schizophrenia vs controls up to 93% accuracy… | ||
| Ben Hamner, CTO, Kaggle, NYC 2017 | Ben Hamner | 21m |
Kaggle reports nearly one million members, says it has realized less than 1% of its potential, and announces acquisition by Google to accelerate growth. It… | ||
| Aaron Roth, Associate Professor, University of Pennsylvania, NYC 2017 | Aaron Roth | 26m |
Shows that concrete privacy failures (AOL, Netflix, genomic reidentification) motivate differential privacy, a formal guarantee where randomized algorithms… | ||
| Alexandra Johnson, Software Engineer, SigOpt, NYC 2017 | Alexandra Johnson | 18m |
Hyperparameter tuning uses SigOpt Bayesian optimization to search real model settings not defaults, yielding concrete wins like a 315% boost in accuracy on a… | ||
| Byron Galbraith, Chief Data Scientist, Talla, NYC 2017 | Byron Galbraith | 17m |
Builds a realtime recommender for a VR silverware shopping service and claims Bayesian bandits find better options faster than A/B tests. Uses Thompson… | ||
| Corinna Cortes, Head of Research, Google, NYC 2017 | Corinna Cortes | 38m |
Describes concrete methods to make neural nets production-ready: AdaNet jointly learns architecture and weights with provable generalization bounds and… | ||
| Claudia Perlich, Chief Scientist, Dstillery, NYC 2017 | Claudia Perlich | 25m |
Models chase the strongest signal and can mislead when data mixes different behaviors. Using logistic regression on an anonymized Facebook-style likes/age set… | ||
| Erik Bernhardsson, CTO, Better Mortgage, NYC 2017 | Erik Bernhardsson | 23m |
Library Annoy builds a forest of random projection trees that returns approximate nearest neighbors much faster than brute force, e.g. a scan that took 2m34s… | ||
| Ross Goodwin, Technologist, Sunspring, NYC 2017 | Ross Goodwin | 20m |
Built a series of working devices that narrate photos and environments using LSTM char-RNN and image-captioning models, training on movie scripts, song lyrics… | ||
| Evan Estola, Lead Machine Learning Engineer, Meetup, NYC 2017 | Evan Estola | 21m |
Meetup recommendation engineering faces extreme cold start sparsity from time and place so simple collaborative filters reproduce historical biases — women… | ||
| Jeff Bradshaw, Founder, Adaptris. NYC 2017 | Jeff Bradshaw | 17m |
Data from farm management systems, tractors, drones, satellites and sensors is ingested into the open source HPCC Systems pipeline, cleaned with ML entity… | ||
| Layla El Asri, Research Scientist, Maluuba, NYC 2017 | Layla El Asri | 31m |
Describes a concrete pipeline and results: a separation of concerns RL architecture splits tasks into subtasks (one agent per subtask) and is tested on a… | ||
| Mayur Thakur, Managing Director, Goldman Sachs, NYC 2017 | Mayur Thakur | 22m |
Built a Hadoop-based platform that flattens diverse firm data into a few canonical tables (one aims for one HBase table for orders) and runs MapReduce… | ||
| Soumith Chintala, AI research engineer, Facebook, NYC 2017 | Soumith Chintala | 22m |
Dynamic Deep Learning: a paradigm shift in AI research and tools:
AI research has seen many shifts in the last few years. We’ve seen research go from using… | ||
| Yi Wang, Tech Lead of AI Platform, Baidu, NYC 2017 | Yi Wang | 22m |
Fault-tolerable Deep Learning on General-purpose Clusters:
Researchers have been used to running deep learning jobs on clusters. In industrial applications… | ||
| Ben Lau, Quantitative Researcher, Hobbyist, NYC 2017 | Ben Lau | 28m |
Builds a self-driving car in the open-source TORCS simulator using a deep reinforcement approach (policy-gradient / actor-critic style) that inputs… | ||
| nothing matches | ||