The thousands of inputs a single neuronal cell receives can interact in complex ways that depend on their spatial arrangement and on the biophysical properties of their respective dendrites. For example, operations such as coincidence detection, pattern recognition, input comparison, and simple logical functions can be carried out locally within and across individual branches of a dendritic tree. In this talk, we will present the hypothesis that the brain leverages these fundamental integrative operations within dendrites to increase the processing power and efficiency of neural computation. We will focus on sensory processing and spatial navigation, with the goal of understanding the mechanistic basis of these brain functions.
We are in the process of transitioning to a new economy where highly complex, custom products are manufactured on demand by automated manufacturing systems. For example, 3D printers are revolutionizing production of metal parts in aerospace, automotive, and medical industries. Manufacturing electronics on flexible substrates opens the door to a whole new range of products for consumer electronics and medical diagnostics. In this talk, I will show that computation is an integral component of modern design and manufacturing. I will demonstrate how computational tools allow creating digital materials with precisely controlled physical properties and how these digital materials are used to automatically synthesize product designs with desired specifications. I will also show how computational tools enable real-time, closed-feedback loop in additive manufacturing systems to improve their reliability and to fabricate complex products with integrated electronics.
Justin Solomon X-Consortium Career Development Associate Professor of Electrical Engineering and Computer Science
Asuman Ozdaglar Joseph F. and Nancy P. Keithley, Professor of Electrical Engineering and Computer Science
The birth of artificial-intelligence research as an autonomous discipline is generally thought to have been the month long Dartmouth Summer Research Project on Artificial Intelligence in 1956, which convened 10 leading electrical engineers — including MIT’s Marvin Minsky and Claude Shannon — to discuss “how to make machines use language” and “form abstractions and concepts.” A decade later, impressed by rapid advances in the design of digital computers, Minsky was emboldened to declare that “within a generation ... the problem of creating ‘artificial intelligence’ will substantially be solved.”
The problem, of course, turned out to be much more difficult than AI’s pioneers had imagined. In recent years, by exploiting machine learning — in which computers learn to perform tasks from sets of training examples — artificial-intelligence researchers have built special-purpose systems that can do things like interpret spoken language or play Atari games or drive cars using vision with great success.
But according to Tomaso Poggio, the Eugene McDermott Professor of Brain Sciences and Human Behavior at MIT, “These recent achievements have, ironically, underscored the limitations of computer science and artificial intelligence. We do not yet understand how the brain gives rise to intelligence, nor do we know how to build machines that are as broadly intelligent as we are.”
Poggio thinks that AI research needs to revive its early ambitions. “It’s time to try again,” he says. “We know much more than we did before about biological brains and how they produce intelligent behavior. We’re now at the point where we can start applying that understanding from neuroscience, cognitive science and computer science to the design of intelligent machines.”
Contact
Assistant