ADVERTISEMENT
  About the SA Blog Network













Guest Blog

Guest Blog


Commentary invited by editors of Scientific American
Guest Blog HomeAboutContact

Ella Gale: Building a Neuromorphic (Brainlike) Computer

The views expressed are those of the author and are not necessarily those of Scientific American.


Email   PrintPrint



Q&A with 6 out of 200 young researchers participating at the 1st Heidelberg Laureate Forum 2013

This is a part of a series of Q&As with mathematicians and computer scientists participating at the 1st Heidelberg Laureate Forum, September 22-27, 2013. More than 40 Laureates (Abel Prize, Fields Medal, Nevanlinna Prize, Turing Award) will attend the forum together with 200 young researchers. For a full week Heidelberg in Germany will be the hot spot of mathematics and computer science. Six of the young scientists told us about their current research and their expectations before the meeting.

Ella Gale

Ella Gale

Meet Ella Gale in this short Q&A series with 6 out of 200 young researchers.

Name?

Ella Gale

Nationality?

British

Where are you based?

I’m currently based in Bristol jointly working at the Bristol Robotics Laboratory and the University of the West of England

What is your current position?

Research Fellow

What is the focus of your research?

I am interested in unconventional computing, which is centres on using unusual phenomena for computation and understanding the natural world as computation. My specific area is memristors. These are novel electronic components that are a consequence of symmetry in electromagnetism. Synapses in the brain (the connection between brain cells through which learning takes place) have been found to work like memristors. Part of the neurons (brain cells) which is involved in neural signal transport can possibly be modelled by memristors.

Thus, I am interested in building a neuromorphic (brain-like) computer from memristors. I think that the lack of success so far in making truly intelligent machines may be because we have so far lacked brain-like components to make it: however, even if memristor networks fail at this task, it would be interesting.

I’m also interested in using bio-inspired approaches in computing as a source of novel algorithms and approaches and understanding how biology fits in within computation theory.

Why did you become a computer scientist?

My degree was in Chemistry, but during it I drifted towards the theoretical end culminating in my master’s project which involved stochastic theory and my first simulation. After that I was hooked, I found it wonderful to be able to make a model and investigate it theoretically using both analytical techniques but especially by doing simulations which allows us to take the model apart and understand it. My PhD followed on from this, I worked using computational simulations to understand chemistry and I was interested in computing at that time because of its usefulness to the physical sciences. For my post-doc I switched from doing computational chemistry to using chemistry to build computers. In this role I was able to go deeper into computing, specifically computability, computer hardware and artificial intelligence.

I think I was attracted to computer science because it is a new field with wide-open vistas and there is a lot of room to be a generalist, in fact a general overview is necessary for computing in a way that it isn’t for the natural sciences. The idea that computational theory can explain life is one I find very interesting and exciting.

Anything like a favorite project?

My favourite project is usually the one I’m currently working on! I’ve enjoyed designing and building logic gates with memristors, because this allowed a lot of creative thinking about what exactly logic is and how it can be implemented. I also enjoyed deriving the analytical theory for memristor operation based on fundamental electromagnetic theory, and it was very gratifying when it worked well.

What about your life beyond research?

I write sci-fi novels, I’ve penned 7 now (of a series) and I’m working on editing them and getting them published. I’m also a keen photographer, especially black and white, infrared, lomo style, landscape, and, due to attending conferences, travel, of course. I also run, swim and practise sword fighting and latin dancing.

Why did you apply for the HLF13?

I have ideas about computing that I am eager to discuss and it seemed like a wonderful opportunity to discuss them with great scientists.

What do you expect from this meeting?

I’m expecting to have many interesting conversations! I’m hoping to learn many interesting things about computer science and mathematics, but I’m also interesting in learning about the job of being an academic scientist from people who have made a big impact in their areas. Perhaps I might meet some future collaborators among the other delegates.

Do you have any Laureates on your list, you would love to talk to?

I tend to prefer serendipity, you can put yourself in the position of meeting people who might inspire you to something, but rarely can you say in advance who or how, so I shall probably try to talk to everyone!

…..

This blog post originates from the official blog of the 1st Heidelberg Laureate Forum (HLF) which takes place September 22 – 27, 2013 in Heidelberg, Germany. 40 Abel, Fields, and Turing Laureates will gather to meet a select group of 200 young researchers. Beatrice Lugger is a member of the HLF blog team. Please find all her postings on the HLF blog.

 

Beatrice Lugger About the Author: Beatrice Lugger is Deputy Scientific Director of the National Institute for Science Communication, Germany. She has a diploma in chemistry and has been working as a science journalist for nearly 20 years for various prestigious German newspapers and magazines. Beatrice is an expert in social media, launched and established Scienceblogs in Germany and writes about science communications on her blog ‘Quantensprung‘. Follow on Twitter @BLugger.

The views expressed are those of the author and are not necessarily those of Scientific American.






Comments 2 Comments

Add Comment
  1. 1. bucketofsquid 5:32 pm 09/17/2013

    Why is the scientist stereotype so nerdy when as this blog post indicates about Dr. Gale, so many of them are artists, writers and athletes as well?

    Link to this
  2. 2. rufusgwarren 11:29 pm 09/17/2013

    I like the idea, but is it necessary to have organic like components. What about a fuzzy machine for each of the senses and recognition. Something like, everything is a member of every set; however, the best fit to a set will define the objects set, and the best fit within the set will define the object. Note that it will also be possible to bias the machine, such as the Asimov rules. With this paradigm, create the fuzzy sets to mimic humans, it should be capable of outperforming humans.

    IBM did something similar using a statistical fit to data. The fuzzy machine idea can be resolved to using mathematical rules to solve problems that are mathematical in nature, the basic knowledge of mankind. Without the present computing power, note that other machines may be designed for a particular problem. Wish I had time for a chart, but I think you get the idea.

    But classical logic can be used as well as the fuzzy logic. Neural networks can be defined from a fuzzy algorithm that are a well defined logic.

    So items not identifiable are either not programmed or is unknown. For these cases a different fuzzy algorithm can be defined to create a set and give it a name with the caveat that if informed by humans, this will define the set and a human must enable the use of this knowledge in order not to cancel any rules.

    One last point, some feedback systems will be unreconciled, hence we might need a new space with imaginary numbers, i.e. typically we would use only 0 through 1, but since its fuzzy, it could be the plane and not a line. I wish I had the knowledge to define a tensor space as the fuzzy set, but I’m afraid I might create impossible scenarios such as Einstein’s GT.

    Also basic control theory will also be applicable as well as all the empirical laws of physics. I tried using a fractal algorithm, i.e. use the same set within each subspace, something like a line with intersecting triangles, and quads. etc. and within each triangle or quad use the same space again, this would scale the response. It was not good for control because an oscillation appeared that mimics the shape of the set. But it might be something that could be used.

    Link to this

Add a Comment
You must sign in or register as a ScientificAmerican.com member to submit a comment.

More from Scientific American

Scientific American Back To School

Back to School Sale!

12 Digital Issues + 4 Years of Archive Access just $19.99

Order Now >

X

Email this Article

X