Imagine a world where machines can see and understand the world the way humans do. Rapid progress in artificial intelligence has led to smartphones that recognize faces, cars that detect pedestrians, and algorithms that suggest diagnoses from clinical images, among many other applications. The success of computer vision is founded on a deep understanding of the neural circuits in the brain responsible for visual processing. This book introduces the neuroscientific study of neuronal computations in visual cortex alongside of the psychological understanding of visual cognition and the burgeoning field of biologically-inspired artificial intelligence. Topics include the neurophysiological investigation of visual cortex, visual illusions, visual disorders, deep convolutional neural networks, machine learning, and generative adversarial networks among others. It is an ideal resource for students and researchers looking to build bridges across different approaches to studying and developing visual systems.
Applications are now open for the 2020 edition of the Brains, Minds and Machines in Woods Hole, MA.
Directors: Gabriel Kreiman, Boston Children’s Hospital, Harvard Medical School; Boris Katz, Massachusetts Institute of Technology; and Tomaso Poggio, Massachusetts Institute of Technology
Location: Marine Biological Laboratory, in Woods Hole, MA.
Course Dates: August 6 – August 27, 2020
Application deadline: April 27, 2020
The basis of intelligence – how the brain produces intelligent behavior and how we may be able to replicate intelligence in machines – is arguably the greatest problem in science and technology. To solve it, we will need to understand how human intelligence emerges from computations in neural circuits, with rigor sufficient to reproduce similar intelligent behavior in machines. Success in this endeavor ultimately will enable us to understand ourselves better, to produce smarter machines, and perhaps even to make ourselves smarter. Today’s AI technologies, such as Watson and Siri, are impressive, but their domain specificity and reliance on vast numbers of labeled examples are obvious limitations; few view this as brain-like or human intelligence. The synergistic combination of cognitive science, neurobiology, engineering, mathematics, and computer science holds the promise to build much more robust and sophisticated algorithms implemented in intelligent machines. The goal of this course is to help produce a community of leaders that is equally knowledgeable in neuroscience, cognitive science, and computer science and will lead the development of true biologically inspired AI.
The class discussions will cover a range of topics, including:
- Neuroscience: neurons and models
- Computational vision
- Biological vision
- Machine learning
- Bayesian inference
- Planning and motor control
- Social cognition
- Inverse problems & well-posedness
- Audition and speech processing
- Natural language processing
These discussions will be complemented in the first week by MathCamps and NeuroCamps, to refresh the necessary background. Throughout the course, students will participate in workshops and tutorials to gain hands-on experience with these topics.
Core presentations will be given jointly by neuroscientists, cognitive scientists, and computer scientists. Lectures will be followed by afternoons of computational labs, with additional evening research seminars. To reinforce the theme of collaboration between (computer science + math) and (neuroscience + cognitive science), exercises and projects often will be performed in teams that combine students with both backgrounds.
The course will culminate with student presentations of their projects. These projects provide the opportunity for students to work closely with the resident faculty, to develop ideas that grow out of the lectures and seminars, and to connect these ideas with problems from the students’ own research at their home institutions.
This course aims to cross-educate computer engineers and neuroscientists; it is appropriate for graduate students, postdocs, and faculty in computer science or neuroscience. Students are expected to have a strong background in one discipline (such as neurobiology, physics, engineering, and mathematics). Our goal is to develop the science and the technology of intelligence and to help train a new generation of scientists that will leverage the progress in neuroscience, cognitive science, and computer science. The course is limited to 35 students.
Longwood Medical Area Ophthalmology Conference.
Gabriel Kreiman presents: Visual Recognition: How it works and what to do when it doesn’t work.
Tang H, Schrimpf M, Lotter W, Moerman C, Paredes A, Ortega Caro J, Hardesty W, Cox D, Kreiman G. (2018) Recurrent computations for visual pattern completion. PNAS, 115:8835-884
Tang H, Kreiman G. (2017). Recognition of occluded objects. In Computational and Cognitive Neuroscience of Vision. (ed Zhao, Q). Singapore: Springer-Verlag
Tang H, Buia C, Madhavan R, Madsen J, Anderson W, Crone N, Kreiman G. (2014). Spatiotemporal dynamics underlying object completion in human ventral visual cortex. Neuron, 83:736-748.
Gabriel Kreiman gives a lecture at the 50th Annual Conference on Information Sciences and Systems (CISS) at Princeton. More information can be found here.
Wu K, Wu E, Kreiman G (2018). Learning scene gist with convolutional neural networks to improve object recognition. IEEE Annual Conference on Information Sciences and Systems (CISS).