If there is anything that is universally true in machine learning, it is that more data is always better. That’s why data scientists always respond to the “How much data do you need?” question with “How much data can you get?”

But sometimes you don’t have much data, because: a) It’s genuinely hard to get; or b) You haven’t been gathering it for very long and it takes time. This is a much more common circumstance than you might think. And you are not alone if you're experiencing this issue. Luckily, and not surprisingly, the machine learning community has come up with numerous approaches for dealing with it. Come sit in on an exclusive conversation with Dr. Tim Oates, professor and practitioner, who will share what appeals to intuition as to why the methods work and provide concrete examples of their application to simple problems.

We will discuss transfer learning, open source datasets and synthetic data, few shot learning, active learning, and semi-supervised learning. By the end of the evening, you should understand what all of these methods are, when they are applicable, and what kinds of results you can expect from using them.

Dr. Tim Oates is Chief Data Scientist at Synaptiq and an Oros Family Professor of Computer Science and Technology in the Department of Computer Science and Electrical Engineering at the University of Maryland Baltimore County. He received a Ph.D. degree from the University of Massachusetts, Amherst in 2001 with a focus on Artificial Intelligence and Machine Learning, and spent a year as a postdoc in the MIT AI Lab. He is an author or co-author of more than 150 peer reviewed papers in AI, ML, and data mining. Dr. Oates served as Chief Scientist for a big data startup in the contact management space, and has consulted in a wide variety of industries, including healthcare, construction, amusement parks, publishing, and social media.

If there is anything that is universally true in machine learning, it is that more data is always better. That’s why data scientists always respond to the “How much data do you need?” question with “How much data can you get?”

But sometimes you don’t have much data, because: a) It’s genuinely hard to get; or b) You haven’t been gathering it for very long and it takes time. This is a much more common circumstance than you might think. And you are not alone if you're experiencing this issue. Luckily, and not surprisingly, the machine learning community has come up with numerous approaches for dealing with it. Come sit in on an exclusive conversation with Dr. Tim Oates, professor and practitioner, who will share what appeals to intuition as to why the methods work and provide concrete examples of their application to simple problems.

We will discuss transfer learning, open source datasets and synthetic data, few shot learning, active learning, and semi-supervised learning. By the end of the evening, you should understand what all of these methods are, when they are applicable, and what kinds of results you can expect from using them.

Dr. Tim Oates is Chief Data Scientist at Synaptiq and an Oros Family Professor of Computer Science and Technology in the Department of Computer Science and Electrical Engineering at the University of Maryland Baltimore County. He received a Ph.D. degree from the University of Massachusetts, Amherst in 2001 with a focus on Artificial Intelligence and Machine Learning, and spent a year as a postdoc in the MIT AI Lab. He is an author or co-author of more than 150 peer reviewed papers in AI, ML, and data mining. Dr. Oates served as Chief Scientist for a big data startup in the contact management space, and has consulted in a wide variety of industries, including healthcare, construction, amusement parks, publishing, and social media.