Understanding Neuronal Networks: Animat project

Understanding Neuronal Networks: Animat project

A New Artificial Intelligence 8 Kevin Warwick Growing Brains

Biological AI Cultured Neural Networks Technical Aspects What does it involve? Where does it stand? Where is it heading? Problems/issues? Using multi-electrode arrays to investigate the computational properties of cultured neuronal networks

Contents Project concept: overview Prior work in this area Infrastructure building

Restriction Evaporation Movement Stimulation Function of the cholinergic system & relevance Findings Ongoing work Future

Why? Why not? Understand memory Alzheimers Disease Understand neural death/plasticity Stroke Regeneration through stem cells extend memory & life Understand basic learning Future robots? Project Concept

Investigate cellular level correlates to higher behavioural processing. How Re-embody a culture of neurones using a robot, enabling it to interact with its environment and so influence future sensory input. Robot with a Biological Brain A closed loop interface between a biological network and a robot Biological

neural network Grown directly on to Multielectrode array Intranet Dimensionalit y reduction, spike train analysis Culture Robot

mapping, Machine learning. Robot running on powered floor Run Down

Neurones from rat embryos Neurones separated using enzymes Laid out on an MEA 2-D Fed 20 mins projections 1 week brain activity

Approach Culture brain cells directly on to a recording surface and reembody the brain within a robotic body. Multi-Electrode Array (MEA) allows recording from 128 electrodes across the entire culture. 200m TiN Electrodes 30m diameter

Neurone Overview How do neurones Culture process sensory process es input input to produce useful behaviours?

Why re-embody using a machine system? Limited sensory input in vivo results in poorly developed and dysfunctional neural circuitry An embodied culture is able to influence its own self. Environmental interaction should result in more meaningful activity than internal self-referencing alone? Advantages over in vivo (already embodied) Non invasive / non destructive recording. Recording from entire structure.

Circuits develop in the presence of test stimuli. Hardware/software overview Other work Steve Potter (Georgia Tech) First simulated animat Ulrich Egert (Freiburg) Hardware prototyping Analytical tool development (MATLAB) Takashi Tateno (Osaka)

Cortical culture characterisation on MEA Shimon Marom (Haifa) Complexity and learning Character isation 1) Create a stable environment 2)

Clean acquired data 3) Characterise spontaneous activity 4) Set up robot culture interface 5)

Test with simple known response mapping 6) Sort data from electrodes to individual units? 7) Develop analysis tools 1) At which point in development?

2) What type of stimulation? Find suitable features to map between culture activity and robot 8) Use computers to automatically train the culture 9) Map connectivity 10) Model / simulate the culture

11) Compare behaviour to model and refine 3) How to gain the cultures attention? 4) Which areas for input / output? 5) How to effectively store memories Can pharmacological manipulation of cholinergic systems answer Some of these questions Infrastructure building:

culture restriction Cell Density % max ASDR (5 min bins) Infrastructure building: evaporation 100 90 80 70

60 50 40 30 20 10 0 0 1 2

3 4 Hours 5 6 7

8 Infrastructure building: evaporation 0.006 0.004 * 0.003 0.002

* g / hour 0.005 0.001 0 * P<0.05 Original Potter Potter Rings - Modified Potter Rings no inlets Rings stimulation Linux based Open Source (GPL2) Hardware driver and GUI available and tested Test, live and user modes

Infrastructure building: simulation A simulated counterpart is useful for many reasons No physical constrictions Faster development More efficient control VRML 3D Model Imported into Webots robot development software Linked with closed loop Ideal experimental platform for RL

Interim summary First 3 years: Stable environment Variability controlled Culture seeding and growth restricted Ability to take accurate, timestamped measures from all systems Long term recordings Full control over stimulation Real and simulated environments

Stability testing: wall avoidance Current work Reinforcement learning and hidden Markov models Functional connectivity maps Plasticity-induced changes and maintenance Observations/Conclusions

Hebbian Learning Sleep time? 100,000 Neurones typical Neurone Specialisation - Functionality Old Age? Information Youtube robot with a rat brain or Kevin Warwick (1 million downloads)

Google as above New Scientist Next Philosophy of Biological AI Contact Information

Web site: www.kevinwarwick.com Email: [email protected] Tel: (44)-1189-318210 Fax: (44)-1189-318220 Professor Kevin Warwick, Department of Cybernetics, University of Reading, Whiteknights, Reading, RG6 6AY,UK

Recently Viewed Presentations

  • Intracellular Compartments and Transport 2 2- Protein secretory

    Intracellular Compartments and Transport 2 2- Protein secretory

    Most proteins that enter the ER are chemically modified there. Disulfide bonds are formed. The disulfide bonds help to stabilize the structure of those proteins that may encounter changes in pH and degradative enzymes outside the cell—either after they are...
  • Instruction execution and ALU - Al-Mustansiriya University

    Instruction execution and ALU - Al-Mustansiriya University

    Arithmetic micro operations. can be implemented in one composite arithmetic circuit. comprised of : full adders and . multiplexers controls which data is fed into input of the adder (suppose B, A represent that inputs (4-bits) ). The output of...
  • Bohr Diagrams

    Bohr Diagrams

    BOHR DIAGRAMS. There is a pattern to the arrangements of electrons in atoms. Atoms have SHELLS of electrons around the nucleus. 2.3 The Periodic table and Atomic Theory. Niels Bohr showed how electrons are arranged in an atom = A...
  • Surgery Quality and Workflow Manager - SQWM Portal

    Surgery Quality and Workflow Manager - SQWM Portal

    Surgery Quality and Workflow Manager. SQWM Software Overview and Information Flow. ... Signing the Brief Op Note causes SQWM to utilize the TIU function to write the note to CPRS. ... Each tab represents a page for logging different kinds...
  • Welcome to the Harry Since 1956 Celebrating Harrys

    Welcome to the Harry Since 1956 Celebrating Harrys

    The St. George Greek Orthodox Church of Lynn 54 South Common Street, Lynn, MA 01902 Thursday April 3rd through Sunday April 6th High School Division Games Start Thursday Night (April 3rd) Teams Expected to Participate: St. George (Lynn) St. Spyridon...
  • Southwest Ohio Region Transfer Summit BIOLOGY Our Faculty

    Southwest Ohio Region Transfer Summit BIOLOGY Our Faculty

    Our Faculty. 15 full-time, tenure-track faculty. 11 with doctoral degrees. 4 with master's degrees. 28 adjunct faculty. 558 sections of biology courses 17-18AY
  • BIM and the Regulatory Process&quot; The Overlooked Element&quot;

    BIM and the Regulatory Process" The Overlooked Element"

    Before BIM, electronic plan review…(ePlan) There is a lot of work to do to prepare for BIM. Make the move from paper to electronic. Systems must be integrated. SunGard Public Sector + ProjectDox = ePlan Review. Residential plan submission is...
  • The Year in Baseball: 1960 - Society for American Baseball ...

    The Year in Baseball: 1960 - Society for American Baseball ...

    LBJ - had just won a landslide re-election in November, 1964 over Barry Goldwater. Had championed the Civil Rights Act in '64 and would push the Voting Rights Bill in '65. Dean Rusk was a Kennedy appointee and is one...