Linkedin
Senior Staff Applied Researcher
Linkedin Oct 2014 - May 2017
Senior Applied Researcher
Linkedin Aug 2013 - Oct 2014
Applied Researcher
University of Chicago Sep 2007 - Jul 2013
Graduate Student
Microsoft Jun 2011 - Sep 2011
Research Intern
Education:
University of Chicago 2007 - 2013
Doctorates, Doctor of Philosophy, Computer Science
Indian Institute of Technology, Delhi 2003 - 2007
Bachelors, Bachelor of Technology, Computer Science
Skills:
Algorithms Machine Learning Python Latex Pattern Recognition Computer Science C++ C Optimization Matlab R Data Mining Statistics Artificial Intelligence Distributed Systems Mathematical Modeling Information Retrieval Applied Mathematics Natural Language Processing Computer Vision Hadoop
- Redmond WA, US Aastha Jain - Sunnyvale CA, US Ankan Saha - San Francisco CA, US Ayan Acharya - Santa Clara CA, US
International Classification:
G06N 3/08 G06N 3/04 G06Q 50/00
Abstract:
Methods, systems, and computer programs are presented for removing bias among users of an online service based on the amount of user's participation in the online service. One method includes operation for pre-training an invite model that provides a first score associated with a user of an online service and for pre-training an adversarial model that provides a second score, the adversarial model having the first score as an input. Further, the method includes training together the invite model and the adversarial model using an adversarial cost function based on the pre-training of the invite model and the adversarial model. The training together is repeated until discrimination of the invite model is below a predetermined threshold. Further, the invite model is utilized to generate the first scores, where the invite model generates the first scores without bias.
Personalizing Online Feed Presentation Using Machine Learning
- Redmond WA, US Timothy Paul Jurka - Redwood City CA, US Ankan Saha - San Francisco CA, US Collin Dang Yen - Saratoga CA, US
International Classification:
G06F 17/24 H04L 29/08 G06F 17/21 G06N 20/00
Abstract:
Techniques for personalizing a user experience for a user of an online service using machine learning are disclosed herein. In some embodiments, a computer system detects a first request by a first computing device of a first user to access content of an online service, identifies at least one content item to display based on the first request, and selects a first presentation template from amongst a plurality of presentation templates based on the at least one content item and an identification of the first user. In some example embodiments, the plurality of presentation templates is stored in a database of the online service, and each one of the plurality of presentation templates is distinct from one another and defines a corresponding manner in which to display the at least one content item.
- Redmond WA, US Shaunak Chatterjee - Sunnyvale CA, US Ankan Saha - San Francisco CA, US
International Classification:
G06Q 50/00 G06F 16/951 H04L 12/58
Abstract:
A computer-implemented method may determine content items regarding a subject to be high demand and sufficient supply, low demand and supply constrained, high demand and supply constrained, or low demand and supply constrained. The computer-implemented method may determine the following: a supply and demand of content items regarding a subject for members, supply demand ratios for the content items regarding the subject for each of the plurality of members, a median supply demand ratio of the supply demand ratios, a total demand for the content items regarding the subject, a median total demand of total demands for the content items regarding subjects for the members, and a median of median supplies demand ratios for the content items regarding the subjects for the members. The method may perform steps to improve demand or supply of a connection network.
Incorporating Contextual Information In Large-Scale Personalized Follow Recommendations
Disclosed herein are techniques for generating contextual follow recommendations. Consistent with embodiments of the present invention, for each of several specific contexts—for example, a member opts to follow another specific member—a set of contextual follow recommendations are pre-computed. Then, in real time, when follow recommendations are being presented to the member, the recommendation system will first make a determination as to whether a member has taken action consistent with any particular context, and if so, a set of pre-computed contextual follow recommendations will be retrieved for possible presentation to the member.
Techniques For Improving Downstream Utility In Making Follow Recommendations
- Redmond WA, US Ankan Saha - San Francisco CA, US Andrew Hatch - Oakland CA, US
International Classification:
G06N 99/00 G06N 5/02
Abstract:
Described herein is a technique to generate and present follow recommendations. During a first stage or phase, training data are obtained by presenting follow recommendations to some randomly selected set of members, and then observing the collective members' responses. Using the training data, first and second predictive machine-learned scoring models are derived—the first scoring model for use in predicting when a member will opt to follow an entity being recommended, and the second scoring model for use in predicting if the member will engage with content presented via a newly formed follow edge. Then, using the scoring models, follow recommendations are derived, scored, and ultimately selected—based on their scores—for presentation to a member.
Populating A User Interface Using Quadratic Constraints
- Redmond WA, US Shaunak Chatterjee - Sunnyvale CA, US Ankan Saha - San Francisco CA, US
International Classification:
G06N 7/00 G06F 9/44 G06F 7/544
Abstract:
A method may include determining a decision space representing a set of content items to be presented on a user interface of a social networking site, the decision space accounting for competing quadratic constraints and interaction effects, estimating the decision space to linearize the competing quadratic constraints, determining, in the estimated decision space and using an objective function, a display probability for each content item in the set of content items, each respective display probability corresponding to a given content item's probability of display in a specific content slot of a plurality of content slots on the user interface; and causing display of the content items with the highest display probabilities.
Constrained Multi-Slot Optimization For Ranking Recommendations
- Sunnyvale CA, US Ankan Saha - San Francisco CA, US Kinjal Basu - Stanford CA, US
International Classification:
G06N 7/00 G06F 17/30 G06Q 50/00
Abstract:
A system, a machine-readable storage medium storing instructions, and a computer-implemented method are described herein are directed to a Content Optimization Engine that determines a display probability for each content item in a set of content items. Each respective display probability corresponds to a given content item's probability of display in a specific content slot of a plurality of content slots in a social network feed of a target member account in a social network service. The Content Optimization Engine calculates a selection probability for each content item in an ordered set of the content items, based on each display probability and a set of interaction effects. The Content Optimization Engine causes display of the ordered set of content items in the target member account's social network feed based on satisfaction of the first and second targets.
Determining Viewer Language Affinity For Multi-Lingual Content In Social Network Feeds
- Mountain View CA, US Ankan Saha - San Francisco CA, US
International Classification:
G06F 17/30 H04L 29/12 H04L 12/58 H04L 29/08
Abstract:
The present disclosure describes various embodiments of methods, systems, and machine-readable mediums which help determine a user's likely affinity for consuming content (such as an article) in a particular language presented (or to be presented) in a heterogeneous feed of a social network.