NICE 2016

NICE 2016 took place at UC Berkeley,

The workshop information was presented on the page https://neuroscience.berkeley.edu/2016-neuro-inspired-computational-elements-workshop/, which is not available any longer. The content is still visible here on the Internet Archive.

Content copied from the internet archive page (as history):

Workshop focus

Conventional, stored program architecture systems are designed for algorithmic and exact calculations.  However, the problems with highest impact involve large, noisy, incomplete, “natural” data sets that do not lend themselves to convenient solutions by current systems.  Our task is to build upon the convergence among neuroscience, microelectronics and computational systems to develop a new architecture designed to handle these natural data sets. The applications and clarification of the value proposition for new neuro-inspired, neuromorphic systems are critical focal points of this workshop.

Learn about past NICE workshops.

Goal

By bringing together researchers from different scientific disciplines and applications, we seek to provide a nucleation point for the development of next generation information processing and computation architectures that go beyond stored program architecture and Moore’s Law limits.

At this workshop, we will:

  • Present applications that are looking for solutions that are beyond the capabilities of current computational systems
  • Highlight technical approaches that are at the early to middle stages of development for new computational systems
  • Identify pathways and resources to accelerate the development of these new systems

Themes

Neuroscience: Sensory information processing in cells and circuits, mechanisms of plasticity, learning and development.

Theory: Theoretical principles of brain information processing, sparse coding, stochastic computing, the role of spikes, Bayesian computing.

Algorithms: Computational synthesis of brain information processing, deep learning.

Platforms/Hardware: Massively parallel neuromorphic hardware architectures, application of commodity systems, novel digital, analog and mixed-signal architectures, application of novel devices.

Applications: Robotics, spatio-temporal pattern detection, causal relations in big data, prediction, approximate computing.

Agenda

Day 1 – March 7, 2016

Sessions – 8:00 am – 6:00 pm
Welcome Reception – 6:00 pm – 7:00 pm

Day 2 – March 8, 2016

Sessions – 8:00 am – 6:00 pm
Dinner – 6:30 pm – 8:30 pm – Claremont Hotel

Day 3 – March 9, 2016

Sessions 8:00 am – 6:00 pm

Agenda

Click here to view all of the presentations (>> NICE 2016 Workshop playlist, UC Berkeley events channel on youtube). You can also download slides from some of the presentations. See links to download in agenda below.

Click here to view abstracts.

Monday, March 7, 2016*Tuesday, March 8, 2016*Wednesday, March 9, 2016
8:00 amWelcome8:00 amWelcome8:00 amWelcome
8:10 amProgrammatic overview:
Thomas Potok (ORNL)
Karlheinz Meier (EU HBP)
Jacob Vogelstein (IARPA)
Bruce Hendrickson (Sandia)
8:15 amSteve Furber (University of Manchester)
SpiNNaker update
8:10 amTheme 3 overview – Approximate Computing
9:00 amThomas Nowotny and Michael Schmuker (University of Sussex) COMPARING NEUROMORPHIC SOLUTIONS IN ACTION 9:00 amConrad James (Sandia)8:15 amIla Fiete (UT Austin)
9:25 amAmir Khosrowshahi (Nervana)
Rethinking computation: a processor architecture for machine intelligence
9:25 amNarayan Srinivasa (Intel)8:45 amJoe Bates (Singular Computing) Practical Approximate Computing
9:50 amBreak9:50 amBreak9:15 amTheme 4 overview – Co-local Computing and Memory
10:10 amChristof Koch (Allen Institute for Brain Science) Understanding Cortex in an Open Access and High Throughput Manner 10:10 amPentti Kanerva (UC Berkeley)
Computing with Hypervectors
9:20 amSurya Ganguli (Stanford)
10:55 amPaul Rhodes (Specific Technologies, Evolved Machines)10:35 amSpeaker unable to attend9:50 amWilliam Dally (NVIDIA, Stanford)
11:20 am3×10 minute lightning talks:
Lloyd Watts (Neocortix) A Platform For Intelligence
Sek Chai (SRI) Deep Temporal Models
Brad Aimone (Sandia) Modulating Neural Computation
11:00 am3×10 minute lightning talks:
Paul Franzon (NCSU)
Katie Schumann (ORNL) Evolutionary Optimization: A Training Method for Neuromorphic Systems (
Mihai Petrovici (Univ. of Heidelberg) Stochastic inference with deterministic spiking neurons
10:20 amBreak
11:55 amLunch11:45 amLunch10:40 amBruce Hendrickson (Sandia) Summary of Conference
1:00 pmDharmendra Modha (IBM)
TrueNorth: Recent Advances in Technology and Ecosystem
1:00 pmRod Rinkus (Neurithmic Systems) Sparse Distributed Representation Trumps Machine Parallelism for Improving Computational Efficiency 11:00 amComputing and Neuroscience Discussion
1:25 pmAlice Parker (USC)
Tradeoffs in Neuromorphic Circuit Design: Reliability, Efficiency, Density of Computations, Power, and Biomimicity
1:25 pmJeff Hawkins (Numenta) What is Intelligence, that a Machine Might Have Some? 12:00 pmLunch
1:50 pmBruno Olshausen (UC Berkeley, Redwood Center for Theoretical Neuroscience) Beyond inspiration: Three lessons from biology on building intelligent machines 1:50 pmGiacomo Indiveri (University of Zurich) Neuromorphic electronic circuits as key enablers of autonomous cognitive agents 1:00 pmMax Di Ventra (UCSD)
2:15 pmPoster and demo overview2:15 pmBreak with posters1:25 pmJesse Engel (Baidu)
2:35 pmBreak with posters2:35 pmTheme 1 overview – Spike Based Representations1:50 pm4×10 minute lightning talks:
Kristofer Bouchard (Lawrence Berkeley National Laboratory)
Tarek Taha (Univ. of Dayton)
William Severa (Sandia)
Dhireesha Kudithipudi (RIT)
3:00 pmDan Hammerstrom (DARPA) UPSIDE / Cortical Processor Study 2:40 pmWolfgang Maass (Technische Universitat Graz)
Principles of network optimization through STDP and rewiring
2:35 pmBreak with posters
3:20 pmKarlheinz Meier (EU HBP) Neuromorphic Computing in the European Human Brain Project 3:10 pmAurel Lazar (Columbia)
NeuroInformation Processing Machines
3:00 pmArvind Kumar and Winfried Wilcke (IBM) 3D Wafer Scale Integration:
A Scaling Path to an Intelligent Machine
3:40 pmKris Bhaskar (KLA/Tencor)3:40 pmTheme 2 overview – Asynchronous Computing3:20 pmGert Cauwenberghs (UCSD) Reverse Engineering the Cognitive Brain in Silicon
4:05 pmChristos Papadmitriou (UC Berkeley)3:45 pmFritz Sommer (UC Berkeley, Redwood Center for Theoretical Neuroscience)3:40 pmJohannes Schemmel (University of Heidelberg) Accelerated Analog Neuromorphic Hardware
4:30 pmPanel discussion/open mic4:15 pmRajit Manohar (Cornell) Asynchronous Logic: A Computer Systems Perspective 4:05 pmGarrett Rose (University of Tennessee – Knoxville) A Memristive Dynamic Adaptive Neural Network Array
6:00 pmDinner at Clark Kerr4:45 pmPanel discussion/open mic4:30 pm – 6:00 pmPanel discussion/open mic
7:00 pm – 10:00 pmPoster and demo session open6:15 pm – 8:00 pmHosted dinner at The Claremont Hotel

Invited Speakers (Confirmed)

Steve Zornetzer, NASA Ames
Steve Furber, University of Manchester
Winfried Wilcke, IBM
Wolfang Maass, Technische Universität Graz
Michael Schmuker, University of Sussex, Brighton, UK
Giacomo Indiveri, Institute of Neuroinformatics, University of Zurich
David Kirk, Nvidia
Alice Parker, USC
Amir Khosrowshahi, Nervana Systems
Bruce Hendrickson, Sandia National Laboratories
Dharmendra Modha, IBM

Organizers

Daniel Hammerstrom, DARPA
Karlheinz Meier, University of Heidelberg
James B. Aimone, Sandia National Laboratories
Jacob Vogelstein, IARPA
Robinson Pino, DOE Office of Science
Paul Rhodes, Evolved Machines
Bruno Olshausen, UC Berkeley, Redwood Neuroscience Institute
Murat Okandan, NICE Workshop Foundation

Sponsors

Helen Wills Neuroscience Institute (2016 workshop host)
Defense Advanced Research Projects Agency (DARPA)
Human Brain Project
Intelligence Advanced Research Projects Activity (IARPA)
Redwood Center for Theoretical Neuroscience
Sandia National Laboratories
Department of Energy Office of Science

Registration

Interested in attending this workshop? All attendees and speakers must register. Please click here to register.*  If you are interested in presenting at the workshop, we ask that you submit an abstract (1 page max pdf) during registration. There is an opportunity to upload abstracts for a talk or the Student Poster Competition at the bottom of the “personal information” page. After your information is entered you will be asked to pay. The 3 ​day ​event ​costs ​$275.00 ​per ​person, ​plus ​processing ​fees.

*If you would like to reserve a spot at the workshop but cannot pay at this time, you can pre-register here.

Additional Info

Parking
Parking at the Clark Kerr Conference center is $10 per day.

Hotels
We have reserved a block of rooms at the Claremont Hotel for $219 per night, plus taxes and fees. Parking at the hotel is an additional $20 per day. You will get more information about how to reserve a room in our block during registration.