Deep Learning from Logged Interventions - Thorsten Joachims
In this talk, I will explore deep learning methods for batch learning from logged bandit feedback (BLBF). Following the inductive principle of Counterfactual Risk Minimization for BLBF, this talk presents an approach to training deep networks from propensity-scored bandit feedback, demonstrating its effectiveness for applications ranging from visual object detection to ad placement.
Видео Deep Learning from Logged Interventions - Thorsten Joachims канала Criteo Eng
Видео Deep Learning from Logged Interventions - Thorsten Joachims канала Criteo Eng
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
![Mingyuan Zhou - Adaptive Diffusion-based Deep Generative Models](https://i.ytimg.com/vi/-TGbGmNpqpk/default.jpg)
![Probabilistic Rank and Reward model - Project 42](https://i.ytimg.com/vi/bMH0hDWE4PE/default.jpg)
![NABD Conf Ann Arbor 2019](https://i.ytimg.com/vi/ep6yBul2Sq8/default.jpg)
![Automatic Backward Filtering Forward Guiding for Markov processes and graphical models](https://i.ytimg.com/vi/XjBO4GSc0i8/default.jpg)
![Jean Pachebat Accelerated Gradient Boosting with Higher Order Optimization Methods](https://i.ytimg.com/vi/UtOM9Ivn1tc/default.jpg)
![DataDisco, One schema to rule them all and kill your data legacy - Fr. Visconte, Math Chataigner](https://i.ytimg.com/vi/0OWw1TEAimI/default.jpg)
![Moving to the Cloud: A Story from the Trenches - Josh Baer](https://i.ytimg.com/vi/I8PF5_AlJdk/default.jpg)
![Where is AI headed?](https://i.ytimg.com/vi/_1LWbVVUsUc/default.jpg)
![Fan Li - Propensity score in Bayesian causal inference: why, why not, and how?](https://i.ytimg.com/vi/_BjkF2nl7dg/default.jpg)
![Not Another Big Data Conference trailer- Paris 2018](https://i.ytimg.com/vi/wyNnL5CBfno/default.jpg)
![Chief Technical Officer Dan Teodosiu - AI Ahead of Us](https://i.ytimg.com/vi/VrULRVElJBU/default.jpg)
![Improving Exploration in Reinforcement Learning: Recent Theoretical Insights - Alessandro Lazaric](https://i.ytimg.com/vi/nQG6YBdA9rg/default.jpg)
![MLiTRW 2016 - Highlights](https://i.ytimg.com/vi/P7JFOJpAuMY/default.jpg)
![The Last Mile: Getting Actionable Insights from Artificial Intelligence - Jason Harper](https://i.ytimg.com/vi/tZlFGRUo9j4/default.jpg)
![Comparative Study of Counterfactual Estimators - Thomas Nedelec](https://i.ytimg.com/vi/xWHIUGGZ7Xs/default.jpg)
![Deep Learning in Practice - Ameet Talwalkar](https://i.ytimg.com/vi/GMY0stioA4g/default.jpg)
![Algorithmic challenges in real-time bidding - Nicolò Cesa Bianchi](https://i.ytimg.com/vi/2P_JEmJXgKQ/default.jpg)
![MLiTRW'18 - The Highlight of the 4th annual Machine Learning in the Real World workshop](https://i.ytimg.com/vi/qhXPmQTUQMk/default.jpg)
![Criteo's privacy preserving ML competition - 1st Place](https://i.ytimg.com/vi/4cS6_Y-dXWA/default.jpg)
![Time Series Jobs Scheduling at Criteo With Cuttle - Guillaume Bort](https://i.ytimg.com/vi/oyOasa-G_f0/default.jpg)