With ODSC West just a few months away, coming up this November 1st-3rd, we know you can’t wait to get the deets on our expert speakers and session topics. Well, your wait is almost over. Although we are still finalizing our lineup and will be adding more speakers in the coming months, we’ve just added several new speakers and session titles to the conference. Check them out below.

Building Production-Ready Recommender Systems with Feast: Danny Chiao | Engineering Lead | Tecton

The challenges of building recommender systems are many and include, frequently changing data, complex data dependencies, and data leakage. In this talk, you’ll learn about these challenges and how to utilize feature stores to alleviate many of the operational challenges that plague recommender systems.

Machine Learning with Python: A Hands-On Introduction: Clinton Brownley, PhD | Data Scientist | Meta

As one of the most powerful programming languages, and because of its many open source machine learning libraries, Python is an essential tool for both newcomers and experienced practitioners alike. In this workshop, you will get hands-on experience in using models for classification and regression, as well as making predictions, in Python. 

Detecting Changes Over Time with Bayesian Change Point Analysis in R: Aric LaBarr, PhD | Associate Professor of Analytics | Institute for Advanced Analytics at NC State University

Instead of using control charts to see if your data have changed after an intervention, you can use a Bayesian approach to examine the whole series and find the point at which the highest probability of change occurred. In this session, you’ll learn how to answer questions such as “Has the financial asset lost value after the recent financial report? Are the healthcare outcomes at this hospital better after our new process to help patients? and Did the manufacturing process improve after upgrading the machinery?” with Bayesian change point analysis in R.

https://odsc.com/california/

Anomaly Detection with Python and R: Luis Vargas, PhD | Partner Technical Advisor | Microsoft

To ensure that your organization is basing its decision-making on accurate data, it’s essential to do the work of anomaly detection. In this session, you’ll explore anomaly detection through the lens of fraud and cover topics such as feature creation and transformation, statistical-based approaches, and learning-based approaches.

Building Modern Search Pipelines with Haystack, Large Language Models, and Hybrid Retrieval: Malte Pietsch | CTO & Co-Founder | deepset

In this session, you’ll learn about recent advancements in semantic search and which developments are worth paying more attention to. You’ll also learn about best practices from use cases and how to utilize Haystack to create and deploy search pipelines. 

Practical Tutorial on Uncertainty and Out-of-distribution Robustness in Deep Learning: Balaji Lakshminarayanan, PhD | Staff Research Scientist | Google Brain

Predictions made by deep neural networks that don’t have well-calibrated predictive uncertainty estimates can make overconfident errors that make trusting their predictions difficult. In this session, you’ll learn about Google Brain’s recent work on creating neural networks that are capable of knowing what they don’t know. 

Large Scale Deep Learning using the High-Performance Computing Library OpenMPI and DeepSpeed: Jennifer Dawn Davis | Staff Field Data Scientist | Domino Data Lab

This session will teach you how the OpenMPI library can be used to train highly complex and massive models. The first part of this session will cover OpenMPI and feature hands-on practice working with it with a Python interface. The second part will provide several examples of how to use DeepSpeed on OpenMPI. It will also cover when to use these tools and when not to. 

MLOps for Deep Learning: Diego Klabjan, PhD | Professor | Northwestern University and Yegna Jambunath | ML Ops Research Specialist | Northwestern University

We know that data is constantly changing and that our models should be updated to reflect those changes. Knowing when and how to train it, however, can be more challenging. In this sessions, you’ll learn how to detect drift and decide when to retrain and how to approach two issues of lifelong training: efficient retraining and catastrophic forgetting. 

Responsible AI Is Not an Option: Scott Zoldi, PhD | Chief Analytics Officer | FICO

In this session, Dr. Zoldi will address the importance of implementing Responsible AI practices in your organization. Not only will he explain why, but he’ll also provide advice on how to identify risks and what tools can be used to mitigate them.

Register for ODSC West 2022 here

As mentioned above, we are still finalizing our speaker lineup for West 2022, so be sure to check our speaker page often for the latest additions. Now is not the time to delay getting your pass, however. You’ll save 60% on any in-person or virtual pass when you register now. But, you’d better act fast. This limited-time offer won’t last long.