To make artificial intelligence more effective and provide users with a secure experience, machine learning safety & responsible AI are critical elements in growing trust between data teams and their end-users. This is why now is a great time to learn more about these growing sub-fields of data science, and how the future can be shaped by them. So check out some sessions at ODSC West this November 1st-3rd, which focus on responsible AI and machine learning safety

Responsible AI is Not an Option

Scott Zoldi, PhD, Chief Analytics Officer at FICO will discuss the importance of AI fairness and reducing bias as the technology becomes more interwoven in the average person’s daily life. Learn why it’s important to use AI ethically and securely while being transparent and keeping customers’ interests at heart. 

Human Factors of Explainable AI

Without explanations, how can data science teams expect stakeholders to trust and adopt machine learning-based technology? Listen in as Meg Kurdziolek, PhD, Sr. UX Researcher at Google, discusses bringing clarity to machine learning technologies for non-expert stakeholders. Discover how end-users interpret explanations to better design your ML systems.

AI in a Minefield: Learning from Poisoned Data

What happens when your well of data is poisoned by malicious actors? Discover the challenges faced by teams faced with dirty data, an overview of data poisoning attacks, and how to reduce the impact of these from any data source with Johnathan Roy Azaria, Data Scientist Tech Lead at Imperva. 

Introduction to Differential Privacy Concepts

As more and more privacy protection regulations are enacted by governments, being able to keep your project in compliance is critical. Learn from Veena B. Mendiratta, PhD, Adjunct Faculty, Network Reliability & Analytics Researcher at Northwestern, as she dives into the topic and how differential privacy concepts can be used in practice. 

Confidential Data Analytics And Learning Data Scientist 

Oftentimes, the data you’re working with is confidential, and protecting the information is of utmost importance for the team. Join Raluca Ada Popa, PhD, Associate Pressor at Berkeley and Co-Founder of PreVeil, as she discussed methods of data encryption that enable teams to run analytics queries on encrypted data.

Quality Control Data Science Life Cycle Through Predictability-Computability-Stability (PCS)

At the core of Data Science, are machine learning & statistics which rely on human judgment at every step of the data life cycle. But these judgment calls come with serious risk for any project depending on the data life cycle, so how do you minimize the risk? Join Bin Yu, PhD, Distinguished Prefffor at the Unversity of California at Berkeley as she discusses the PCS framework. Learn how it can expand unify, expand, and streamlines ideas and best practices of machine learning & statistics. 

Cybersecurity and Policing in the Metaverse

As virtual spaces such as the Metaverse grow, so does a user’s ability to participate in a market that runs on virtual assets. These assets are the centerpiece of commerce within these spaces and require rules & regulations that protect the market place which they are held. Listen in as Jack McCauley, Board Trustee at the University of California at Berkeley discusses policing and security within this virtual landscape. 

Emerging Approaches to AI Governance: Tech-Led Vs. Policy Led

As AI failures and missteps occur, these issues will only become more sensitive as ai-powered programs continue to grow in people’s lives. It’s not the trust in the systems that are at risk, but forms of liability. In this session, you’ll learn from Ilana Golbin, Director at PwC Emerging Technologies & Responsible AI Lead, as she explores the pros and cons of tech-led AI governance and policy-led AI governance. 

Robust and Equitable Uncertainty Estimation

Anyone or anything can make a prediction, so how can we trust the predictions made by machine learning models when there’s a lack of standards or historical information to back up an ML model’s prediction? Join in, and learn from Aaron Roth, PhD, Professor of Computer and Cognitive Science at the University of Pennsylvania. He’ll discuss a new technique that addresses these problems and methods to produce prediction sets for arbitrary black-box prediction. 

Register for ODSC West 2022

There you have it! The latest in machine learning safety & responsible AI is all waiting for you at ODSC West. Join us and learn from the experts as they go into depth on these topics and so much more. If you buy a ticket now, you’ll save 20% off the ticket price. And remember, ODSC also provides a world-class virtual experience so you’ll get the most from the conference without the stresses of travel. Register today!