Leif H. Finkel - US grants

9313526 Buchsbaum This award allows the University of Pennsylvania to renovate and modernize research space in order to enhance it's Neuroengineering Research Laboratories. By modernizing these facilities, this university will be able to integrate research and training in engineering technology with the study of brain function. While emphasizing vision, the integration of psychology, physiology, computer science and engineering technology can lead to tremendous advances in research in such areas as machine vision at this institution. The research is lead by 15 highly regarded faculty, while 4 postdoctoral fellows, 43 graduate students and 42 undergraduate students will take advantage of the renovated facility. The University of Pennsylvania has a unique environment for this activity since it has an NSF Science and Technology Center in Cognitive Science along with the department of neuroengineering and an Institute in Neurological Sciences. The Neuroengineering Research Laboratories will provide a central facility for collaborative research, and will offer the capability for advanced training in state-of-the-art simulation methods in neural networks. Finally, this department interacts with Lincoln University (an HBCU) in order to include more minority participation in the engineering activities at the University of Pennsylvania. ***

IBN: 9634367 PI: Finkel Gain control is a powerful process that is used in many biological and engineering applications. In a videocamera, the photodetectors can only operate over a limited range of light levels, so in order to be able to record in bright daylight and dim internal light, the signal must be re-scaled to stay within a limited range of intensities. A slightly smarter system, used in more expensive devices, is to change the gain (how much the signal is amplified) separately in each local region. Changing the amount of gain in an adaptive manner allows signals to be extracted from ambient noise. The same principles of gain control act in our visual system-- in a simple manner on the pupil to control light entering the eye, but in increasingly more sophisticated ways as one goes to higher levels in the visual cortex. Our goal is to investigate these mechanisms of cortical gain control to better understand how the visual sytem works, and as a means of uncovering new image processing applications. The cortex has the ability to control gain not only based on spatial location, but also in a context-dependent manner. The response to a particular feature, say a short line segment, depends upon what other features are present in the image. If the line segment forms part of a circle or an extended line, the response is increased. Psychophysical studies have identified the conditions under which certain features "pop-out" and become the focus of visual attention. Recent physiological studies have shown that cortical cells are exquisitely sensitive to small changes in the context of the scene. It is believed that response changes are a result of a change in gain of individual cortical cells, but the mechanisms controlling cell gain are poorly understood. We are particularly interested in differential changes of gain among different cells in a neural population. Most features are represented, in the cortex, by a set of graded responses ove r a population of cells. The orientation of a line, for example, is coded by a distribution of responses over cells that prefer various orientations (vertical, horizontal, oblique). We hypothesize that inputs from stimuli in the visual image act to differentially change the gain of cells. A differential change in gain across a population would lead to increased sensitivity--a better estimate of line orientation, or whatever feature is being analyzed. This type of context-dependent gain control could lead to enhanced discrimination of the salient features in an image. We will carry out a series of physiological experiments in cat visual cortex to investigate this hypothesis. We also will conduct a set of computer simulations of various cellular mechanisms to test our model. Finally, we will build a prototype image processing device, based on these findings, and apply it to real images.

9873463
Finkel
We Propose an approach to the study of intelligent systems based on the modeling and construction of neuromorphic knowledge systems. Neuroinorphic systems offer a platform for integrating insights from mathematics, neurobiology, and Computer simulation, and represent a new type of neural-based computing. Our approach arises from the premise that the traditional components of intelligence (including Sy requires a new technology; this technology will be provided by together as an integrated system. The study of such complex learning, recognition, and attention) are Properties of a common underlying neural structure, and that these compo ents are best studied neuromorphic systems engineering. Ncuromorphic VLSI chips stems and systems use subthreshold analog circuits to model dendritic computation and asynchronous digital circuits to model axonal communication . These systems can function autonomously or be interfaced with conventional computers. Neuromozphic systems offer the possibility of learning and recognition on real-time, real-world problems such as visual search and the recognition of biological motion.
Our approach is based on spike-based computation: use of the precise timing of ncuronal spikes for representation and computation. Spike-based computations underlie some of the most impressive behaviors in biology, from bat sonar to human vision. In particular, we will focus on neuronal synchronization and will develop models in which learning, attention, and recognition all operate through effects on synchronization. Such models require neural mechanisms for the detection of synchronization (Recognition), enhancement of synchronization (Learning), and dynamic modulation of synchronization (Attention).
We will use a team approach in which faculty and students work together on linked projects. The four investigators have expertise in complementary areas: neuromorphic chip design (Boahen), cellular biophysics and neural network simulation (Finkel), physiological data analysis and modeling (Gerstein), theoretical analysis of learning and spikebascd computation (Hopficid). Together we %ill investigate mechanisms for the control of synchronization based on changes in neuromodulation, on selective modification of
AMPA/NMDA conductance ratios, on temporallyasyminetric synaptic plasticity, and on synaptic delay and spike-frequency adaptation. These simulations @l be tied to analysis of synchronization in spike-train recordings from awake, behaving animals engaged in learning, attention, or recognition tasks. Analysis of physiological and psychophysical data will motivate development of theoretical models of spike-based computation. Most critically, development of a series of neuromorphic chips will be carried out as a means of scaling up the simulations to systems containing over 200,000 spiking neurons each making thousands of connections. To facilitate development and sharing of ideas, we will develop a common simulation environment for analysis of spiking neurons that will allow an interface between large-scale network simulation, neuromorphic chip design, and analysis and prediction of physiological recordings. This enviroranent will be made available (via the Web) for widcr use by physiologists and modelers, and as an educational tool.
Our current retina-based neuromorphic chips (Boahen, 1998) outperform CCD technology as a result of ncurally-derived mechanisms. These chips incorporate biological mechanisms for local gain control and motion detection. Prototype systems of interconnected spikegenerating chips will allow real-time feedback control of connectivity. We outline a plan for interfacing cortically-based ncuromorphic systems with high-speed multiprocessors to allow real-time attention, recognition, and learning on continuous video input streams. Recent ncurophysiological studies suggest that neuronal responses in alert animals viewing rcal-iiorld environments differ radically from responses to simplified stimuli. Such results argue for the study of neural-based systems interacting with real-world environments.
The outcome of these studies will be tools, technology and insights that provide a new domain of interaction between neuroscientists, computer scientists and mathematicians and engineers. They will yield insight into the spike-based computations underlying intelligent behavior. The technology developed will have immediate application to robotics and intelligent neural prostheses.

[unreadable] DESCRIPTION (provided by applicant): We propose an interdisciplinary program at the University of Pennsylvania to train outstanding undergraduate and predoctoral students in computational neuroscience. Penn offers unique strengths in basic and clinical neuroscience, both integrated with computational approaches. All participating departments at Penn are national leaders in their field, have a long history of research and educational collaborations, and are located in close proximity on a single campus. The undergraduate and graduate student populations are highly qualified. Twenty-four faculty at Penn, plus faculty at six nearby regional institutions are involved, including experimentalists, modelers, and many faculty with expertise in both domains. The preceptors have extensive experience in education and research training in computational neuroscience. The program consists of three components: an undergraduate research training program, a summer research program for undergraduates, and a predoctoral training program. The focus is on directly integrating Neuroscience and quantitative studies through course work and extensive research training. Students will carry out integrated experimental/modeling research projects directed at computational problems. The structure and strategy of our program is designed to have each student individually achieve a significant research contribution through the integrated development of model, experiment, and data analysis. We propose to develop an integrated undergraduate curriculum in computational neuroscience including a new, keystone course, and to significantly update and expand our current graduate courses in computational neuroscience-all including substantial laboratory components. We will also introduce a dedicated seminar series, separate undergraduate and graduate journal clubs, an annual retreat, and other programmatic activities. A distinguishing focus of our program is on application of computational neuroscience to neurological and psychiatric disorders. Students will undertake clinical rotations, analyze clinically obtained data, and have the option of rotations on computational projects in Penn's clinically-directed research centers. A summer research program will be developed, which will attract undergraduates primarily from the Philadelphia region. Six nearby universities are participating in this summer program, including Swarthmore, Drexel, Temple, Haverford, Bryn Mawr, and Lincoln Universities. The summer program will be designed to engage and excite students to pursue graduate work in computational neuroscience. The student population will include a significant proportion of women and underrepresented minorities. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): We propose an interdisciplinary program at the University of Pennsylvania to train outstanding undergraduate and predoctoral students in computational neuroscience. Penn offers unique strengths in basic and clinical neuroscience, both integrated with computational approaches. All participating departments at Penn are national leaders in their field, have a long history of research and educational collaborations, and are located in close proximity on a single campus. The undergraduate and graduate student populations are highly qualified. Twenty-four faculty at Penn, plus faculty at six nearby regional institutions are involved, including experimentalists, modelers, and many faculty with expertise in both domains. The preceptors have extensive experience in education and research training in computational neuroscience. The program consists of three components: an undergraduate research training program, a summer research program for undergraduates, and a predoctoral training program. The focus is on directly integrating Neuroscience and quantitative studies through course work and extensive research training. Students will carry out integrated experimental/modeling research projects directed at computational problems. The structure and strategy of our program is designed to have each student individually achieve a significant research contribution through the integrated development of model, experiment, and data analysis. We propose to develop an integrated undergraduate curriculum in computational neuroscience including a new, keystone course, and to significantly update and expand our current graduate courses in computational neuroscience-all including substantial laboratory components. We will also introduce a dedicated seminar series, separate undergraduate and graduate journal clubs, an annual retreat, and other programmatic activities. A distinguishing focus of our program is on application of computational neuroscience to neurological and psychiatric disorders. Students will undertake clinical rotations, analyze clinically obtained data, and have the option of rotations on computational projects in Penn's clinically-directed research centers. A summer research program will be developed, which will attract undergraduates primarily from the Philadelphia region. Six nearby universities are participating in this summer program, including Swarthmore, Drexel, Temple, Haverford, Bryn Mawr, and Lincoln Universities. The summer program will be designed to engage and excite students to pursue graduate work in computational neuroscience. The student population will include a significant proportion of women and underrepresented minorities. [unreadable]