Regina Barzilay is a professor at MIT and a world-class researcher in natural language processing and applications of deep learning to chemistry and oncology, or the use of deep learning for early diagnosis, prevention and treatment of cancer. She has also been recognized for her teaching of several successful AI-related courses at MIT, including the popular Introduction to Machine Learning course. This conversation is part of the Artificial Intelligence podcast. If you would like to get more information about this podcast go to https://lexfridman.com/ai or connect with @lexfridman on Twitter, LinkedIn, Facebook, Medium, or YouTube where you can watch the video versions of these conversations. If you enjoy the podcast, please rate it 5 stars on iTunes or support it on Patreon.
The book talks about the development of cancer treatments and drugs while also diving into the author's life transformation as she moves to a new country and adapts to a new culture.
Discover the history behind drug development and the importance of prioritizing goals that matter to you, not just the scientific community.
Over 1.7 million cancer cases are detected in the US each year, leading to 600,000 related deaths, and computer scientists believe that advancements in machine learning can aid in the prediction of various types of cancer and improve treatment effectiveness.
Building large healthcare data sets for research purposes has been hindered by issues related to trust and transparency in data access. Despite the availability of de-identified data, legal and ethical issues surrounding ownership and responsibility of patient data pose challenges to the development of large-scale data sets for medical research.
This podcast discusses the important role of incentives in driving adoption and innovation in the medical imaging field. The conversation includes a case study on the use of machine learning to identify breast cancer risk based on image density.
This podcast episode highlights the increasing importance of drug design in healthcare, especially with the ability to generate molecules with desired properties for in silico screening. The speaker also emphasizes that while diagnostics and imaging take ideas from other areas, drug design is a technically interesting and exciting field.
The speaker discusses how linguistic approaches to drug development have been limited in their successes compared to models that utilize other sources such as domain experts and combinatorial spaces of molecules or graphs.
The limitations of machine learning include both the data problem and the way we train our systems. However, it is possible that human connection and language can be improved using machine learning in the future.
This podcast discusses the potential for brain-altering technology to change the way we perceive and interact with the world, as well as the impact it may have on our motivation and attention.
The speaker reflects on the universality of vanity and the importance of finding alone time and reading to reconnect with oneself. They also mention the potential of feedback loops in areas beyond video games, such as relationships and personal missions.