From scaling machine learning to solving the sparsity of data using computer vision, we have collated five video presentations from the Applied AI Virtual Summit which took place last week. Hear from eBay, LinkedIn, Facebook, AI Reverie and the World Economic Forum below.

Training Models at Facebook Scale with PyTorch

Mohamed Fawzy, Senior Engineering Manager & Tech Lead, Facebook & Dwarak Rajagopal, Senior Engineering Manager & Tech Lead, Facebook

Large scale distributed training has become an essential element to scaling the productivity for ML engineers. Today, ML models are getting larger and more complex in terms of compute and memory requirements. The amount of data Facebook train on is huge. In this talk, you will learn about the Distributed Training Platform to support large scale data and model parallelism. Mohamed & Dwarak touch base on Distributed Training support for PyTorch and how they are offering a flexible training platform for ML engineers to increase their productivity at Facebook scale.

You can see the full presentation here.

Automated Model Exploration

Sandeep Jha, Staff Technical Program Manager, LinkedIn & Jun Jia, Senior Staff Software Engineer, LinkedIn

Machine Learning modeling work has a lot of repetitive steps, and LinkedIn can automate some of those to speed up the path to make an impact. In this talk, Sandeep & Jun provide a brief overview of LinkedIn's initiative on ML model exploration and dive deep into automating standard model training practices such as feature selection, hyperparameter tuning, and model selection.

You can watch the full video here.

Why AI Ethics Matter

Kay Firth-Butterfield, Head of AI and Machine Learning and Member of the Executive Committee at World Economic Forum

In Kay's role as Chief officer of the EAP, she has advised governments, think tanks and non-profits about AI. Kay is also the co-founder of the Consortium for Law and Policy of Artificial Intelligence and Robotics at the Robert E. Strauss Center & teacher at the University of Texas. Kay's presentation surrounded the world of ethics in AI and why it is so important that set guidelines are adhered to.

You can watch the full video here.

Solving Hard Problems in Computer Vision With Synthetic Data

Daeil Kim, Co-Founder & CEO, AI.Reverie

CEO of AI Reverie, Daeil Kim,  gave a talk on solving hard problems in Computer Vision with Synthetic Data. AI Reverie make the case that a necessary step towards the advancement of computer vision will be to solve the bottlenecks of data curation and annotation. During this presentation, Daeil presents how synthetic data can solve both of these issues while providing additional advantages over real-world data. The talk also concluded by discussing real-world case studies for orbital insights, retail, and agriculture that are currently being solved using synthetic data.

Watch the full presentation here.

Machine Learning: Build, Deploy, Monitor at Scale

Tom Virant, Director, Data Science, eBay

How do you deploy and manage not just 1, but 100’s of machine learning models? What changes when you move from batch processing to synchronous calls? When your algorithms become mission critical, what changes? What is seemingly easy when working for a small company, with lower volumes or with batch operations becomes exponentially more difficult at scale. Off the shelf development and deployment platforms have existed for years, but use cases are becoming broader and corporations are starting to customize and invest internally. In this presentation, Tom discuss the infrastructure and processes necessary to operationalize data science at scale.

View the full video on youtube here.