Скачать или смотреть (Keynote) NLP AutoML best practices and Working with Multilingual Data

Timestamp:
00:00 Topic Introduction
01:00 Agenda
02:05 Workflow of ML Process
05:01 NLP Tasks with AutoML
07:49 MultiLingual NLP
10:28 #1 Translation
12:34 #2 Character Embeddings
14:33 #3 MultiLingual Embeddings
16:50 #4 MultiLingual Models
18:42 #4 Transfer Learning
20:33 H20.ai Approach for NLP with AutoML
22:10 NLP with AutoML Transformers Pipeline
23:39 NLP with AutoML AutoModel Pipeline
24:12 NLP with AutoML Custom Pipeline

This event is brought to you by DataScience SG, a group for anyone interested in data science to interact and share their skills and know-how.

Find out more about DataScienceSG at goo.gle/DataScienceSG.

Topic: NLP AutoML best practices and Working with Multilingual Data
Presenter: Shivam Bansal (Senior Data Scientist, H2O.ai)

We come across many languages while working in APAC region, and multilingual NLP is key to extract unified insights out of multilingual data. While working with text data in English is challenging, building and applying NLP models for multiple languages adds to the complexity of this task.

In this talk, we will introduce the approach developed by Kaggle Grandmasters at H2O.ai to leverage multilingual transformer based models and transfer learning to solve NLP for multilingual data. We will also conduct a hands-on session using H2O’s Hybrid AI Cloud, which will cover -
1. AutoML Best Practices
2. NLP Best Practices
3. Expert settings for multilingual NLP and transfer learning

Shivam Bansal is a Senior Data Scientist, Kaggle grandmaster, and a multi-year winner of kaggle analytics competitions and awards. Shivam specializes in building end-to-end data science and machine learning applications solutions.

He has a blend of full-stack engineering, core data science, and business skills. He has helped customers in the areas of banking, finance, insurance, retail, and healthcare to enable a data science-driven business transformation. He has extensive hands-on industrial experience in building models, algorithms, and engines from scratch and has led data science teams to deliver end-to-end production-ready data-driven products. Shivam holds a Bachelor's in Computer Science and a Master's degree in data science from the National University of Singapore and was the Valedictorian of his cohort.

*

For more updates on upcoming events, follow us on social media:
✉️ Newsletter → https://goo.gle/devspace-news
👤 Facebook →   / devspacesg  
🐦 Twitter →   / devspacesg  
🔴 Meetup →https://www.meetup.com/developer-space/