Chris' Thesis Page
Here are some BSc and MSc thesis topics I'm interested in supervising.
Feel free to e-mail for more of a discussion.
- I'm interested in spiking neural networks, and have a mechanism for
learning to categorise with them. There's
a page with
code and a paper .
- I was working on the Human Brain Project .
- I'm flexible, so if you have a neural idea, ask.
- Here are some ideas:
- Spiking Neural Systems
- Categorising with neurons (CANT, nest, spiNNaker)
- Parameter set a 1 ms. FLIF model. Change my 10ms
Fatiguing Leaky Integrate and Fire neuron model to a 1ms model,
and validate it on biological data.
- Chatbot on Nest
- Chatbot on spiNNaker.
- An agent using the neurorobotic platform.
- Improved vision system in spiNNaker or Nest.
- Introduce a new neural model to Nest, spiNNaker or both.
- Introduce a new synapse model to Nest, spiNNaker or both.
- Any Neuro-Cognitive Model.
- Any rule based system in neurons (we have a translation
mechanism)(easy).
- Implement interactive activation theory in neurons (Nest).
Sharks and Jets in Nest. This might be a nice way to
provide an alternative to our CA model that is immediately
relevant to cognitive models.
- Natural Language Processing Systems
- Chatbot
- Chatbot for Pandemic
- GPT-3 for generation
- Plink Parser for New Sentences
- Tutorial Agent
- I'm interested in tutorial agents for aspects of CST 3170.
- Other AI Systems
- Chess
- Pandemic Agent
- I'm interested in modelling the spread of diseases (like Covid-19).
- General: I've superevised a lot of theses over the years using
the
University of California at Irvine's Categorisation
Benchmark . The idea is that a a student uses a machine
learning algorithm on tests on the benchmark. Any algorithm will
work. You can use multiple benchmarks. You can use multiple
algorithms. It's good to use an N-Fold test.
- General: A chatbot.
- Very General: Anything in AI
This page was developed by
Chris Huyck
Page last modified