Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
Effective Machine Learning Needs Leadership — Not AI Hype
 Originally published in BigThink, Feb 12, 2024.  Excerpted from The...
Today’s AI Won’t Radically Transform Society, But It’s Already Reshaping Business
 Originally published in Fast Company, Jan 5, 2024. Eric...
Calculating Customer Potential with Share of Wallet
 No question about it: We, as consumers have our...
A University Curriculum Supplement to Teach a Business Framework for ML Deployment
    In 2023, as a visiting analytics professor...
SHARE THIS:

4 years ago
Wise Practitioner – Predictive Analytics Interview Series: Vishal Hawa at The Vanguard Group

 

By: Luba Gloukhova, Founding Chair, Deep Learning World

In anticipation of his upcoming conference presentation at Deep Learning World
May 31-June 4, 2020, we asked Vishal Hawa, Principal Scientist at The Vanguard Group, a few questions about their deployment of predictive analytics. Catch a glimpse of his presentation, Multi-channel Optimal Path Sequencing Through Bayesian Deep Learning, and see what’s in store at the Deep Learning World conference in Las Vegas

Q: In your work with deep learning, what do you model (i.e., what is the dependent variable, the behavior or outcome your models predict)?

A: One of the applications that we use deep learning is in Marketing cost optimization. We predict conversion rates on various pathways taken by leads. So, our dependent variable becomes conversion (binary) and cost to acquire a customer.

Q: How does deep learning deliver value at your organization what is one specific way in which model outputs actively drive decisions or operations?

A: In Marketing optimization we have to deal with numerous combinations of treatments/campaigns. The DL model picks up potentially viable combinations from almost infinite possibilities.

Q: Can you describe a quantitative result, such as the performance of your model or the ROI of the model deployment initiative?

A: It’s a very noisy field and I would be happy if the model performance beats 85%, with heavy tuning and crafting features we are able to perform better than 90% accuracy. From our simulation studies it will provide cost savings of about 30%.

Q: What surprising discovery or insight have you unearthed in your data?

A: Not every campaign treats customers uniformly, the impact of campaign treatments depends on type of customer and time of life cycle. Deep Learning model finesses the treatment based on attributes of the leads.

Q: What excites you most about the field of deep learning today?

A: The biggest gain I find from Deep Learning is opening up the space with multiple architectures which can then be mixed to create hybrid architectures each focused to deliver in specialized use cases from generative simulations to anomaly detections.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Deep Learning World.

A: Given a numerous (say almost infinite combinations) you will be able to filter out the ones that the potential for maximum lift or benefits.

—————————–

Don’t miss Vishal’s presentation, Multi-channel Optimal Path Sequencing Through Bayesian Deep Learning, at Deep Learning World on Wednesday, June 3, 2020 from 11:20 AM to 12:05 PM. Click here to register for attendance.

By: Luba Gloukhova, Founding Chair, Deep Learning World

This content is restricted to site members. If you are an existing user, please log in on the right (desktop) or below (mobile). If not, register today and gain free access to original content and industry news. See the details here.

Comments are closed.