Maintaining oil and gas machinery is expensive—but predictive maintenance models can help engineers minimize repairs and downtime.

Shayan Mortazavi and Alex Lowden, Data Scientists at Accenture in the Industrial Analytics Group, work on the development of predictive maintenance models to minimize downtime of systems. In this episode, they discuss the complications when building these models, such as limited access to failure data and the massive number of features available, as well as the need for explainability and interpretability in their models. They also share how SigOpt’s parallelism feature allowed them to accelerate model development.

• 1:27 - Intros
• 3:05 - Machinery maintenance, then vs now
• 4:06 - Goals of maintenance
• 6:49 - Challenges of predictive maintenance for oil and gas
• 8:31 - Human in the loop element
• 10:07 - Interpretability
• 11:42 - Using SigOpt to optimize hyperparameters
• 13:50 - Managing multiple LSTMs
• 16:38 - Using SigOpt's multimetric optimization
• 18:36 - Predicting ultimate machine failure
• 20:39 - Getting teams on board with AI-based tools
• 23:21 - Overconfidence of AI 

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt

Learn more about Accenture: https://www.accenture.com

Subscribe to our YouTube channel to watch Experiment Exchange interviews: https://www.youtube.com/channel/sigopt

How do you design a better glass?

Paul Leu, Associate Professor of Industrial Engineering at the University of Pittsburgh, shares the insights behind his interdisciplinary work using machine learning to design and test novel glass structures. He and Michael McCourt discuss the collaboration between Pitt and SigOpt, the challenges of glass design and testing, and what's ahead for the Year of Glass.

• 0:23 - Intros
• 5:14 - Paul's interdisciplinary philosophy
• 7:04 - Nanomaterials research
• 11:48 - Designing for the properties of light
• 14:35 - How Paul and his team use SigOpt
• 17:50 - Designing better solar panels
• 21:05 - How Paul uses SigOpt's advanced features
• 22:51 - The Year of Glass
• 24:16 - Paul's work with MDS-Rely 

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt

Learn more about LAMP: https://lamp.pitt.edu/ 

Learn more about MDS-Rely: https://mds-rely.org/ 

Learn more about the Year of Glass: https://www.iyog2022.org/ 

Subscribe to our YouTube channel to watch Experiment Exchange interviews: https://www.youtube.com/channel/sigopt

Given a property, what’s the material or the molecule that achieves it?

This is the question behind some of Rafael Gomez-Bombarelli's latest work. Tune in as SigOpt's Michael McCourt interviews the Assistant Professor of Materials Processing at MIT about his development of machine learning strategies to design new materials—including fluids, cloths, metals, and nanomaterials. 

• 1:28 - Intros
• 2:18 – Using ML to predict physical properties of molecules
• 3:48 – Rafael's active learning process
• 5:23 - ML costs
• 6:34 – How Rafael uses SigOpt in his work
• 7:59 - Modeling the color change properties of molecules
• 9:26 - Multi-fidelity methods
• 12:12 – Continuum simulations
• 14:57 – Inverse design problems
• 18:50 - Workshops at ML conferences
• 20:47 – What's next for Rafael

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt 

Subscribe to our YouTube channel to watch Experiment Exchange interviews: https://www.youtube.com/channel/sigopt

Aluminum design is an incredibly complicated business. Not only do you have to get the model design right—it also has to work in the real world. In fact, the aluminum soda can is one of the most engineered products in your house right now. In this episode, SigOpt’s Head of Engineering Michael McCourt talks with Vishwanath Hegadekatte, R&D Manager at Novelis, about how he's using tools like SigOpt to optimize aluminum production and design—as well as considering environmental impact to build better products and conserve resources. 

• 2:16 - Intros 
• 4:03 - Metrics for aluminum design 
• 4:50 - How Novelis is pioneering physics-informed machine learning
• 6:40 - How Novelis uses SigOpt
• 8:35 – Industry is on par with academia for physics-informed machine learning 
• 12:18 - "The optimum is not always the best choice" 
• 14:45 - What's next from Novelis

Learn more about Novelis: https://www.novelis.com 

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt

Subscribe to our YouTube channel to watch Experiment Exchange interviews https://www.youtube.com/channel/sigopt

How do you detect fraud when less than one percent of your network’s users are bad actors? In this episode, SigOpt’s Head of Engineering Michael McCourt speaks with Venkatesh Ramanathan, a Director of Data Science at PayPal, about his work using Graph Neural Networks to detect fraud across large financial networks.

• 0:23 - Intro
• 3:08 - AI/ML at AOL
• 4:24 - The scale of data today 
• 6:11 – The tradeoffs of accuracy and interpretability 
• 7:54 - What are Graph Neural Networks? 
• 9:18 - Robustness of GNNs; how they work with blockchain networks 
• 10:57 - The need for robust hardware for GNNs 
• 12:44 - How PayPal uses SigOpt for hyperparameter search 
• 15:12 - The importance of sample efficiency 
• 16:51 - What's next for Data Science at PayPal 
• 20:52 - Opportunities for academia to power industry insights

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt Subscribe to our YouTube channel to watch Experiment Exchange interviews: https://www.youtube.com/channel/sigopt

Machine learning holds significant promise for fields like proteomics, therapeutics, and more—but blockers like access to datasets and issues of health privacy make progress complicated. Alexander Johansen is a Ph.D. student in computer science studying the intersection between computer science, bioinformatics and digital health.

Within his lab at Stanford University, Alexander has applied Natural Language Processing to proteins, explored the history of wearables data privacy, and more. In this episode, SigOpt’s Head of Engineering Michael McCourt speaks with Alexander about his pioneering work and how SigOpt has played a role in advancing progress.

• 1:29 - Intro 2:09 - What Alexander's lab works on
• 5:40 - Who makes up Alexander's team
• 8:39 - Should only the FDA be involved?
• 12:31 - NLP for proteomics 
• 17:44 - How Alexander uses SigOpt to power his research
• 21:24 - How Alexander collaborates with other areas of study
• 24:28 - Picking the right health data to prioritize 

Follow Alexander on Twitter: https://twitter.com/AlexRoseJo 

Learn more about the Stanford Center for Personalized Health: https://stanford-health.github.io/ 

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt 

Subscribe to our YouTube channel to watch Experiment Exchange interviews: https://www.youtube.com/channel/sigopt

Small to medium enterprises make up the majority of the companies in the world, yet they're often underserved when it comes to AI. Pablo Zegers, Co-Founder and VP of Product, is seeking to change this through Anastasia AI, which democratizes access to time series analysis for a variety of businesses. In this episode, they discuss this work, the environmental impacts of today's models, and what's next for Anastasia AI. 

• 2:19 - The state of AI in Latin America
• 9:14 - Why Anastasia is focusing on time series forecasting first in a modular system 
• 11:15 - How SigOpt helps Anastasia get better results for customers 
• 15:00 - Working with X-RNN models 
• 18:43 - What's next for Anastasia AI this year 
• 19:48 - Environmental impact of training AI 

Learn more about Anastasia AI at https://anastasia.ai 

Read the IEEE Spectrum Magazine article about ImageNet energy usage: https://spectrum.ieee.org/deep-learning-computational-cost 

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt 

Subscribe to our YouTube channel to watch Experiment Exchange interviews: https://www.youtube.com/channel/sigopt

Our brains only use about 30-40 watts of power, yet are more powerful than neural networks which take extensive amounts of energy to run. So what can we learn from the brain to help us build better neural networks? Join Michael McCourt as he interviews Subutai Ahmad, VP of Research at Numenta, about his latest work.

In this episode, they discuss sparsity, bioinspiration, and how Numenta is using SigOpt to help them build better neural networks and save on training costs.

1:31 - Background on Numenta
2:31 - Bioinspiration
3:47 - Numenta's three research areas
4:06 - What is sparsity and how does it function in the brain?
7:15 - Training costs, Moore's Law, and how deep learning systems are on a different curve
9:58 - Mismatch between hardware and algorithms today in deep learning
11:04 - Improving energy usage and speed with sparse networks
14:10 - Sparse networks work with different hyperparameter regimes than dense networks
14:18 - How Numenta uses SigOpt Multimetric optimization
15:48 - How Numenta uses SigOpt Multitask to constrain costs
18:06 - How Numenta chose their hyperparameters
19:40 - What's next from Numenta

Learn more about Numenta at numenta.com and follow them on YouTube at www.youtube.com/c/NumentaTheory

Read Jeff Hawkin's book, A Thousand Brains: A New Theory of Intelligence

Learn more about SigOpt at sigopt.com and follow us on Twitter at twitter.com/sigopt

Subscribe to our YouTube channel to watch Experiment Exchange interviews at www.youtube.com/channel/sigopt