Over the past few years, we’ve sought out and spoken to some of the top researchers, entrepreneurs, and executives involved in the machine learning field across industries and domains. This week, we decided to collate our five most popular machine learning podcast episodes in one place (they’re listed below, ranked in order from most popular by number of downloads).
Machine Learning Rising
Since its initial resurgence just over a decade ago, there has been an increasing interest by the public in machine learning as a tangible technology. A diverse array of new learning algorithms, more and faster computing power for a fraction of the cost, and availability of ‘big data’ (within which machine learning systems can apply a granular focus and also learn at the individual level) has positioned this technology as an invaluable augmentation of human capabilities.
Machine learning approaches have infiltrated almost every industry, with software for computer vision, speech and voice recognition, natural language processing, robot control, data mining, and other applications affecting our daily lives. Archival medical records are used to determine the best treatment for individual patients; patterns in longitudinal traffic data are applied to improve traffic control; and sets of big data from experiments are collated and used to further advances in the empirical sciences, from biology to astronomy, and improving causal inferences in the social sciences.
The experts featured below are respected professors, consultants, researchers and entrepreneurs. They have, in one way or another, impacted the machine learning field through pioneering research work or by contributing significant knowledge capital and applying technologies in new and improved ways. Each podcast offers up ideas from a historical, present, and futuristic standpoint, and provides the expert-specific insights that are necessary for anyone truly interested in learning the potential applications and implications of machine and deep learning.
Top Machine Learning Podcast Interviews
Dr. Yoshua Bengio is a key figurehead in machine learning and deep learning circles, dedicated to academia. He co-organized the Learning Workshop with Facebook’s Yann LeCun, with whom he also created the International Conference on Representation Learning (ICLR). Bengio was also an initial point of contact for OpenAI when they sought out the best machine learning and AI researchers in the world.
Bengio was a driving influence in the resurgence of interest in AI and machine learning around 2005; he and others founded a research program funded by CIFAR, or the Canadian Institute for Advanced Research, which focused on deep learning and discovered ways to transition the neural nets of old to the deep learning networks of today.
In this podcast episode, Bengio discusses the proliferation of neural networks and researchers’ progress in tackling some of today’s most pressing challenges – unsupervised learning and learning from unstructured data. Bengio also discusses future applications and expresses great interest in further applying machine learning to the medical domain: “I feel like this is the kind of application where we can directly help people, not just by building better gadgets but by directly influencing the health of humans.”
Dr. Roman V. Yampolskiy is a distinguished tenured associate professor at University of Louisville, founder and director of the university’s Cyber Security Lab, and also author of several well-received books – including Artificial Superintelligence: A Futuristic Approach. In this episode, he speaks on the topic of cyber security, which continues to gain attention as cyber attacks increase and gain more media coverage, from communications to the financial industry to retail customer info. Yampolskiy also discusses the high level of security risk due to automated intelligent systems; at present, AI is being used to help defend systems, but could it become a liability – or a direct threat – in the future?
Machine learning and artificial intelligence methods got a head-start in the world of finance in the 1990s, and interest continues to grow. Tad Slaff is the founder of Inovance Financial Technologies and the creator of TRAIDE, a strategy creation platform that uses machine learning algorithms to help traders uncover patterns in assets and indicators and build more reliable trading strategies. Slaff (and arguably others like) him are trying to leverage what individual humans are good at (creativity) and combine their innovation with machine learning strengths (finding patterns and information in large data sets).
In this podcast episode, Slaff discusses the likely possibility of robo-advisors or brokers becoming an integral part of trading and investing futures, as well as the obstacles in convincing the individual investor to trust such a system i.e. the need for rational transparency at how systems arrive at specific conclusions.
With the contributions of big players like Baidu, Google, Nuance and others, NLP continues to make advances and receive plenty of media coverage. Dr. Dan Roth, founder professor of Engineering at University of Illinois at Urbana-Champaign,;has been published broadly in the fields of natural language processing, learning and inference, and other areas. His significant contributions to the field have earned him fellowship in multiple reputable organizations, including the Association for Advancement of Artificial Intelligence (AAAI) and the Association for Computing Machinery (ACM).
In this podcast episode, Roth talks about NLP progress over the past decade, how we got to where we are today, and the future implications – for example, a near-future time when we’re able to communicate with computers in a more natural way (we’re already seeing developments in smart conversational bots in customer service and children’s toys). He also discusses why the medical and compliance domains continue to be challenging areas and the need to collate research and publish advances in an ‘intelligent’ way so that the latest scientific findings are accessible to doctors, researchers, and others around the world.
Businesses know there’s a lot of hype around the potential insights that can be gleaned from applying machine learning to their existing data, but many are not sure how it works or where to start. That’s where today’s niche machine learning consultants are making strides. Dr. Charles Martin is a freelance consultant in machine learning, data science, and software development. He has worked on machine learning (ML) systems running in production at companies like GoDaddy, Aardvark (Google), eBay, and Blackrock, and he used machine learning to help Demand Media become the first $1B IPO since Google.
In this episode, Martin sheds light on how machine learning got its start in business and enterprise in the 1990s, and how those pioneering companies helped advance the machine learning trends of today. He also speaks on how we can “even the playing field” in machine learning, so that more businesses have access to reliable machine learning systems without having to reinvent the wheel.
Related TechEmergence Interviews/Articles:
- What is Machine Learning?
- Machine Learning Misconceptions – Infographic from our Expert Consensus
- Machine Learning Healthcare Applications – 2016 and Beyond
Image credit: CSO Online