Hello current and future DataScientists!
From the speaker(Bradley Voytek),
One of the most remarkable aspects of data science is the ability to turn observations into quantities, and from those quantities infer physical laws and make predictions about the future.
Drawing on my experiences from both my time in industry as well as from running a neuroscience research lab, in this talk I will examine the art of parameterization: turning large datasets into human-level understanding. For example, neuroscience is creating data at an unprecedented rate. There is an explosion of ‘‘big data’’ initiatives in neuroscience, ranging from recording from thousands of neurons simultaneously, to massive repositories of thousands of peoples' worth of human functional and structural brain imaging, to population-wide genetic data. Despite this wealth of data, major "theories of the brain" remain elusive. This has led many to lament that neuroscience is data rich, but theory poor. Here, I argue that through principled aggregation of these large, diverse, heterogeneous neuroscience datasets, we can address critical issues in cognition and mental health. These data include textual data from the millions of peer-reviewed neuroscience publications, functional and structural brain imaging data, demographic and genetic information, neural electrophysiology, gene expression, and cellular physiology. Through such aggregation and integration, we can mine the data to find missing links and gaps in our knowledge; we can algorithmically generate plausible hypotheses for us. That is, we leverage the wealth of neuroscience data to reduce the paucity of neuroscience theories.
Bradley Voytek is an Associate Professor in Cognitive Science, Data Science, and the Neurosciences Graduate Program at UC San Diego. He is an Alfred P. Sloan Neuroscience Research Fellow and National Academies Kavli Fellow, and a founding faculty member of the UC San Diego Data Science major and the Halıcıoğlu Data Science Institute. He was also the first Data Scientist at Uber in 2011 where he helped build their Data Science team. His lab studies neural oscillations by combining large-scale data mining and machine learning techniques with hypothesis-driven experimentation. He is also a “zombie brain expert” along with friend and fellow neuroscientist Timothy Verstynen, with whom he cowrote the book Do Zombies Dream of Undead Sheep? by Princeton University Press.