Entity Embeddings & PyTorch in the Enterprise

Deep Learning NYC
Deep Learning NYC
Public group

Every 1st Thursday of the month

Location image of event venue


Notice: Change of venue! We're meeting on 34th St, not the Loft. You must register to attend: https://bit.ly/2019-06-06-DL-Meetup

Entity Embeddings

Entity Embeddings is an popular technique for applying deep learning to tabular data. It involves representing the categorical data of an information systems entity with multiple dimensions.

It is used in several production systems at companies such as Google, Instacart, Twitter, etc. In Mahesh Khatri's presentation, he presents the concept along with examining its usage in the following 3 papers:

Kaggle Competition winner papers:
"Artificial Neural Networks Applied to Taxi Destination Prediction", Yoshua Bengio’s team - 31 July 2015,
"Entity Embeddings of Categorical Variables", Cheng Guo, Felix Berkhahn - 22 April 2016, and
"Deep Neural Networks for YouTube Recommendations", Google AI - 16 September 2016.

Mahesh has previously spoken on this topic on https://twimlai.com/­

PyTorch in the Enterprise

Enterprises have different concerns as opposed to research and academia. Principally, research has a focus on data exploration and training models, enterprise has production focus: adaptation with larger solutions, robustness, CI/CD, Compliance: GDPR, PCI, HIPPA, CCPA, monitoring and analytics and more.

In this presentation Partner Solutions Architect and Global Machine Learning Segment leader Kris Skrinak shares his experience working with PyTorch in dozens of organization on AWS. In the process you'll learn about how PyTorch is integrated into Amazon SageMaker, DockerHub as well as ECR, SageMaker Local and open source stack. CI/CD with open source and commercial tools are covered as well.

Every month the deep learning community of New York gathers at the AWS loft to share discoveries and achievements and describe new techniques.