INE Telluride Workshop 2006
Workshop Highlights for 2006
This year’s three week summer workshop included background lectures (from leading researchers in biological, computational and engineering sciences), practical tutorials (from state-of-the-art practitioners), hands-on projects (involving established researchers and newcomers/students), and special interest discussion groups (proposed by the workshop participants). There were 65 attendees, composed of 9 organizers, 6 administrative and technical staff members, 22 invited speakers, and 28 applicants/ students/invitees ine-web.org/telluride-conference-2006/telluride-overview/index.html . Participants were encouraged to become involved in as many of these activities, as interest and time permitted. Two daily lectures (1.5 hrs each) covered issues important to the community in general, presenting the established scientific background for the areas and providing some of the most recent results. These lectures spanned most of the diverse disciplines comprising neuromorphic engineering. Furthermore, additional lectures were presented by group work participants in the afternoons on topic germane to the projects. Participants were free to explore any topic, choosing among the 9 workgroups and approximately 30 hands-on projects, and 6 interest/discussion groups. See ine-web.org/telluride-conference-2006/workgroups-lecturesx/index.html for more details.
The participants were drawn from academia (86%), government laboratories (3%), and industry (10%), although many of our academic participants are involved with or have founded start-up companies even though they have maintained their academic affiliation. To our knowledge, 7% of the academic participants fall in this category. Given the youth of the field, it is not surprising that most of the participants are from academia. Now that we are making inroads into industry, we hope to have a larger industrial contingent in the future, albeit from within. The diverse backgrounds of the participants span medicine (6%), biology (6%), computational neuroscience (9%), neurophysiology (3%), engineering (67%), computer science (3%), and robotics (5%). Of these participants, 56% are from US organizations, 35% from Europe, 9% from Far-East, South America, Oceana and Africa, 16% (down from 27%) are women and 10% (up from 5% to our judgment) are minorities.
The workshop provided background material in basic cellular and systems neuroscience as well as practical tutorials covering all aspects of analog VLSI design, simulation, layout, and testing. In addition, sensorimotor integration and active vision systems were also covered. Working groups were established in robotics (focused on the use of Koalas, LEGO and custom designed legged robots), locomotion (covering central pattern generators and sensory-motor interactions), configurable neural systems (covering multi- chip systems interconnection and computation with large numbers of siliconneurons), active vision (covering stereo vision and visual motion detection), audition (covering speech detection, localization and classification), proprioception (covering the multi- sensory fusion and behaviors), attention and selection, emerging technologies, and finally industrial applications (interaction with Iguana Robotics, Inc., Yahoo, and AFRL). Throughout the three weeks we held morning lectures on all aspects of neuromorphic engineering (see the workshop schedule in Chapter 2). In the first week, in addition to the morning lectures, a group of tutorial lectures covered basic neurophysiology, the vertebrate auditory system, central pattern generator theory, transistors, simple circuit design, an introductions to the programming environments needed for the active vision and audition systems (introduced by the various course participants), and introduction to the Koalas and bipedal robots. In the second week participants focused on workgroup projects, ranging from modeling of the auditory cortex, to working with roving robots, to designing asynchronous VLSI circuits. In the third week, participants focused on the completion of their hands-on projects, collecting data and analysis of their results, while attending lectures, discussion groups, and documenting their work. The Telluride 2006 report contains summaries of all their hard work.
This year we continued our efforts of generating more sophisticated behaving neuromorphic systems by integrating physical and theoretical results from various labs around the world. This Workshop is the ideal venue at which this type of integration can occur because the researchers bring their research set-ups to Telluride; it would be a logistical nightmare for labs to organize such a meeting on their own. Furthermore, the spontaneous collaboration that usually emerges out of this Workshop would not happen without it. We hold this aspect of our Workshop to be the most important because it provides a convergent location where leading researchers in the field meet every year. Projects carried out by subgroups during the workshops included active vision, audition, sonar, olfaction, motor control, central pattern generators, robotics, multi-chip communication, analog VLSI and learning.