Back to All Events

Exploring Automated Modeling with H2O’s Driverless AI w/Pramit Choudhary Lead Data Scientist

  • Hub101 31416 Agoura Rd #105, Westlake Village United States (map)

As the adoption of ML in solving business use-cases has increased, there are often multiple unknowns that Machine Learning Scientist struggle with. To be effective in practice, it would be great to have an automated modeling engine which could help with feature engineering, hyper parameter tuning and model selection within a fixed computational budget (defined by Accuracy, Time and Interpretability).

In this hands-on session, we will explore the usefulness of auto-modeling using real datasets related to supervised learning problems(Classification/Regression/TimeSeries). We will also discuss on how we can further explore and validate these automated models using model diagnostics and interpretation by enabling,

1. Model Validation: Ways to explore and validate black box ML systems enabling model comparison both globally and locally - identifying biases in the training data through interpretation.

2. Interpretable Models: Ability to build natively interpretable models - with the goal to simplify complex models to enable better understanding.

3. What-if Analysis: An interactive environment where communication can happen i.e. enable learning through interactions. User having the ability to conduct "What-If" analysis - effect of single or multiple features and their interactions

4. Model Debugging: Ways to analyze the misbehavior of the model by exploring counterfactual examples(adversarial examples and training)

About our Speaker

Our speaker, Pramit Choudhary, is an Applied Machine Learning Research Scientist/Engineer and currently Lead Data Scientist His area of interest is building scalable Statistical/Machine Learning models(Bayesian and Frequentist Modeling techniques) to help businesses realize their data-driven goals.

Recently, he has been exploring better ways to understand and explain model's learned decision policies to reduce the chaos in building effective models to close the gap between a prototype and operationalized model(prescriptive Machine Learning). In his past life he has worked on ML problems related to NLP (e.g. topic modeling), improving operation efficiency in Oil and Gas Industry (e.g. time series/ anomaly detection), social media analysis, personalized recommendation engines, match-making and fraud detection to name a few

Earlier Event: March 11
Female Entrepreneurship Meetup
Later Event: March 18
Pitch to Your Peers - Conejo Valley