Multi-Stage Retrieval and Ranking [Siddharth Sharma, Facebook]
Speaker: Siddharth Sharma, Facebook
Host: Nishant Sinha, OffNote Labs
In this talk we would broadly cover indexing and ranking techniques used during online ads ranking, e-commerce search, music/image session recommendation etc. The audience would also get a glimpse into the working of scalable machine learning engineering pipelines.
------
The OffNote Labs AI Talk Series brings you industry experts, researchers and practitioners, passionate to share their learnings and experience -- on innovating and building cutting-edge AI technology / systems which touch and influence the lives of billions of people on this planet.
Our talks are informal, a blend of traditional presentation / podcast, and the audience very technically engaged.
We take delight in unraveling the experience of research and innovation, and celebrate innovators who 'follow the problem', and make complex technology work in the wild.
==
Follow OffNote Labs on LinkedIn. / offnote
Watch previous talks and Subscribe on Youtube: https://bit.ly/31VTMHH
Sign up to our newsletter: http://offnote.substack.com
Read our research articles: / offnote
Web: https://offnote.co
===
00:00 Setting Up
15:16 Offnote Labs - Introduction
18:20 Overview - Multi-Stage Ranking & Retrieval
21:40 Introduction - Siddharth Sharma
32:46 Latency Constraints
37:30 Retrieval I0I
41:06 Boolean Retrieval
42:30 Product Attributes
47:46 Bird’s Eye View - Querry: Product Ranking
1:11:06 AD Retrieval & Ranking
1:21:23 Two Tower Model
1:33:03 Light Weight Interaction Model
1:44:43 Deep Structured Semantic Model (DSSM)
1:46:23 Process Flow
1:48:53 Questions - Part I
2:00:50 Search in Embedding Space
2:02:13 Spotify Use Case
2:03:36 Popular Embedding Libraries
2:09:35 Conclusion and Questions
---
#informationretrieval #ranking #learningtorank #search #cascadedsearch #cascade #ads #neuralnetwork #machinelearning #deeplearning #developers #pytorch #machinelearner #matrixmultiplication #deeplearningpython #Offnote #Latency #latencyconstraints #Retrieval #Ranking #Mutistageranking #booleanretrieval #productattributes #twotowermodel
#lightweightinteractionmodel #semanticmodel #dssm #embeddingspace #embeddinglibraries #offnotetalks