ACM CIKM October 31, 2013 Jeff Hawkins On-line Learning From Streaming Data.

Slides:



Advertisements
Similar presentations
From Cortical Microcircuits to Machine Intelligence
Advertisements

ONE STOP THE TOTAL SERVICE SOLUTION FOR REMOTE DEVICE MANAGMENT.
Hierarchical Temporal Memory (HTM)
CHANGING THE WAY IT WORKS Cloud Computing 4/6/2015 Presented by S.Ganesh ( )
© Chinese University, CSE Dept. Software Engineering / Software Engineering Topic 1: Software Engineering: A Preview Your Name: ____________________.
Sparse Coding in Sparse Winner networks Janusz A. Starzyk 1, Yinyin Liu 1, David Vogel 2 1 School of Electrical Engineering & Computer Science Ohio University,
Blue brain Copyright © cs-tutorial.com.
Amazon. Cloud computing also known as on-demand computing or utility computing. Similar to other utility providers like electric, water, and natural gas,
Principles of Personalisation of Service Discovery Electronics and Computer Science, University of Southampton myGrid UK e-Science Project Juri Papay,
1 NETE4631 Cloud deployment models and migration Lecture Notes #4.
© 2013 IBM Corporation October 4, 2013 IT Analytics and Big Data IBM Solutions Paul Smith (Smitty) Service Management Architect.
XProtect® Expert 2013 Product presentation
CloudSocial Mobility Big data Social connections, mobility, cloud delivery and pervasive information are converging in a powerful way. This convergence.
Hybrid Pipeline Structure for Self-Organizing Learning Array Yinyin Liu 1, Ding Mingwei 2, Janusz A. Starzyk 1, 1 School of Electrical Engineering & Computer.
1 SWE Introduction to Software Engineering Lecture 22 – Architectural Design (Chapter 13)
Computer Science 162 Section 1 CS162 Teaching Staff.
Physical Symbol System Hypothesis
Knowledge Portals and Knowledge Management Tools
By: Raj Akula. Professor: Wei Hao. Course: CSC 599. Semester: Fall 2011.
LLNL-PRES This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344.
Department of Computer and Information Science, School of Science, IUPUI Dale Roberts, Lecturer Computer Science, IUPUI CSCI.
Facebook (stylized facebook) is a Social Networking System and website launched in February 2004, operated and privately owned by Facebook, Inc. As.
1 CSC 1401 S1 Computer Programming I Hamid Harroud School of Science and Engineering, Akhawayn University
Operating System Review September 10, 2012Introduction to Computer Security ©2004 Matt Bishop Slide #1-1.
Michael Ernst, page 1 Collaborative Learning for Security and Repair in Application Communities Performers: MIT and Determina Michael Ernst MIT Computer.
Graph Algorithms for Irregular, Unstructured Data John Feo Center for Adaptive Supercomputing Software Pacific Northwest National Laboratory July, 2010.
LAB CVP 2009 ‘Leveraging the LIMS Investment’. Invested in a Laboratory Information Management System (LIMS) Solution is limited to Storing and Reporting.
Advances in Modeling Neocortex and its impact on machine intelligence Jeff Hawkins Numenta Inc. VS265 Neural Computation December 2, 2010 Documentation.
© 2008 IBM Corporation ® IBM Cognos Business Viewpoint Miguel Garcia - Solutions Architect.
COMS E Cloud Computing and Data Center Networking Sambit Sahu
Practical Heirarchical Temporal Memory for Time Series Prediction
(Business) Process Centric Exchanges
A New Theory of Neocortex and Its Implications for Machine Intelligence TTI/Vanguard, All that Data February 9, 2005 Jeff Hawkins Director The Redwood.
The Future of the iPlant Cyberinfrastructure: Coming Attractions.
Chapter 1 — Computer Abstractions and Technology — 1 The Computer Revolution Progress in computer technology – Underpinned by Moore’s Law Makes novel applications.
Computer Organization & Assembly Language © by DR. M. Amer.
IBM Bluemix Ecosystem Development Hands on Workshop Section 1 - Overview.
Introduction To BlueMix By: Ryan
Information Services Internet Accounting Workshop 1 st Feb 2011 Peter Kurtz Manager, Network & Communication Services.
 Programming - the process of creating computer programs.
Chapter 8 – Cloud Computing
Why Can't A Computer Be More Like A Brain?. Outline Introduction Turning Test HTM ◦ A. Theory ◦ B. Applications & Limits Conclusion.
Web Technologies Lecture 13 Introduction to cloud computing.
1 TCS Confidential. 2 Objective : In this session we will be able to learn:  What is Cloud Computing?  Characteristics  Cloud Flavors  Cloud Deployment.
COMP1321 Digital Infrastructure Richard Henson March 2016.
Best SMS Gateway Software Provider Company in India By Aruhat Technologies.
HIERARCHICAL TEMPORAL MEMORY WHY CANT COMPUTERS BE MORE LIKE THE BRAIN?
AMSA TO 4 Advanced Technology for Sensor Clouds 09 May 2012 Anabas Inc. Indiana University.
IPv6 Adoption Status and Scheduling for Sustainable Development 24 July 2012 Nate Davis Chief Operating Officer, ARIN.
The Neural Code Baktash Babadi SCS, IPM Fall 2004.
Introduction to Computers - Hardware
Bhakthi Liyanage SQL Saturday Atlanta 15 July 2017
Connected Maintenance Solution
Cloud computing-The Future Technologies
Overview – SOE PatchTT November 2015.
Overview – SOE PatchTT December 2013.
Sales & Marketing Funnels
Implementing a sequence machine using spiking neurons
Connected Maintenance Solution
Software Design and Architecture
Java programming lecture one
Maintaining software solutions
Datamining : Refers to extracting or mining knowledge from large amounts of data Applications : Market Analysis Fraud Detection Customer Retention Production.
Introduction to Cloud Computing
System And Application Software
Cloud computing mechanisms
Big Data Young Lee BUS 550.
Technical Capabilities
About Thetus Thetus develops knowledge discovery and modeling infrastructure software for customers who: Have high value data that does not neatly fit.
Open Data Cube Demo and FDA Interfaces
Presentation transcript:

ACM CIKM October 31, 2013 Jeff Hawkins On-line Learning From Streaming Data

1) Discover operating principles of neocortex 2)Build systems based on these principles Anatomy, Physiology Theoretical principles Software Industrial Research Track Anomaly detection in high velocity data Cortical algorithms

The neocortex is a memory system. data stream retina cochlea somatic The neocortex learns a model from sensory data - predictions - anomalies - actions The neocortex learns a sensory-motor model of the world

Principles of Neocortical Function retina cochlea somatic 1) On-line learning from streaming data data stream

Principles of Neocortical Function 2) Hierarchy of memory regions retina cochlea somatic 1) On-line learning from streaming data data stream

Principles of Neocortical Function 2) Hierarchy of memory regions retina cochlea somatic 3) Sequence memory - inference - motor data stream 1) On-line learning from streaming data

Principles of Neocortical Function 4) Sparse Distributed Representations 2) Hierarchy of memory regions retina cochlea somatic 3) Sequence memory data stream 1) On-line learning from streaming data

Principles of Neocortical Function retina cochlea somatic data stream 2) Hierarchy of memory regions 3) Sequence memory 5) All regions are sensory and motor 4) Sparse Distributed Representations Motor 1) On-line learning from streaming data

Principles of Neocortical Function retina cochlea somatic data stream x x x x x x xxxxxx x 2) Hierarchy of memory regions 3) Sequence memory 5) All regions are sensory and motor 6) Attention 4) Sparse Distributed Representations 1) On-line learning from streaming data

Principles of Neocortical Function 4) Sparse Distributed Representations 2) Hierarchy of memory regions retina cochlea somatic 3) Sequence memory 5) All regions are sensory and motor 6) Attention data stream 1) On-line learning from streaming data These six principles are necessary and sufficient for biological and machine intelligence. - All mammals from mouse to human have them

Sparse Distributed Representations (SDRs) Many bits (thousands) Few 1’s mostly 0’s Example: 2,000 bits, 2% active Each bit has semantic meaning Meaning of each bit is learned, not assigned …………01000 Dense Representations Few bits (8 to 128) All combinations of 1’s and 0’s Example: 8 bit ASCII Individual bits have no inherent meaning Representation is assigned by programmer = m

A Few SDR Properties 1) Similarity: shared bits = semantic similarity subsampling is OK Indices 1 2 | 10 2) Store and Compare: store indices of active bits Indices | 40

Coincidence detectors How does a layer of neurons learn sequences? Sequence Memory (for inference and motor)

Each cell is one bit in our Sparse Distributed Representation SDRs are formed via a local competition between cells.

SDR (time =1)

SDR (time =2)

Cell forms connections to subsample of previously active cells. Predicts its own future activity.

Multiple Predictions Can Occur at Once With one cell per column, 1 st order memory We need a high order memory

High Order Sequence Memory Enabled by Columns of Cells Cortical Learning Algorithm (CLA) Distributed sequence memory High order High capacity Multiple simultaneous predictions Semantic generalization

1)NuPIC Open Source Project Three Current Directions

Single source tree (used by GROK) GPLv 3 Steady community growth – 67 contributors (+26 since July) – 245 mailing list subscribers – 1621 total messages eBook from community member OS community joining Kaggle Competitions Fall Hackathon: 70 attendees NuPIC Open Source Project

1)NuPIC Open Source Project 2)Custom CLA Hardware -Needed for scaling research and commercial applications -DARPA “Cortical Processor” -IBM, Seagate, Sandia Labs 3)Commercialization Three Current Directions

2. Look at data3. Build models Problem: - Doesn’t scale with velocity and #models Solution: - Automated model creation - Continuous learning - Temporal inference 1. Store data Stream data Automated model creation Continuous learning Temporal inference Predictions Anomalies Actions Past Future Data: Past and Future

Anomaly Detection Using Predictive Cortical Models Cortical Memory Encoder SDR Prediction Point anomaly score Time average Distribution of averages Metric anomaly score Metric 1 Cortical Memory Encoder SDR Prediction Point anomaly score Time average Distribution of averages Metric anomaly score SDR Metric N System Anomaly Score

Largely predictable Largely unpredictable Metric value Anomaly score Metric value Anomaly score

 Smartphone-centric  Ranks anomalous instances  Rapid drill down  Continuously updated  User-controlled notifications Breakthrough Science for Anomaly Detection Reinventing UX for IT Monitoring Grok for IT Monitoring  Detects problems thresholds miss  Continuous learning  Automated model building  State-of-the art neocortical model In private beta for Amazon AWS cloud users

 Custom metrics for any application/server  Web interface and mobile client source code available under no-cost license  Engine API to be published  NuPIC open source community Extensible Architecture