Projects per year
Most approaches for optimisation of neural networks are based on variants of back-propagation. This requires the network to be time invariant and differentiable; neural networks with dynamics are thus generally outside the scope of these methods. Biological neural circuits are highly dynamic yet clearly able to support learning. We propose a reinforcement learning approach inspired by the mechanisms and dynamics of biological synapses. The network weights undergo spontaneous fluctuations, and a reward signal modulates the centre and amplitude of fluctuations to converge to a desired network behaviour. We test the new learning rule on a 2D bipedal walking simulation, using a control system that combines a recurrent neural network, a bio-inspired central pattern generator layer and proportional-integral control, and demonstrate the first successful solution to this benchmark task.
|Title of host publication||2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2018)|
|Place of Publication||Madrid, Spain|
|Publisher||Institute of Electrical and Electronics Engineers (IEEE)|
|Number of pages||6|
|Publication status||Published - 7 Jan 2019|
|Event||2018 IEEE/RSJ International Conference on Intelligent Robots and Systems - Madrid, Spain|
Duration: 1 Oct 2018 → 5 Oct 2018
|Conference||2018 IEEE/RSJ International Conference on Intelligent Robots and Systems|
|Abbreviated title||IROS 2018|
|Period||1/10/18 → 5/10/18|
FingerprintDive into the research topics of 'A Bio-inspired Reinforcement Learning Rule to Optimise Dynamical Neural Networks for Robot Control'. Together they form a unique fingerprint.
- 1 Finished
1/01/14 → 31/12/16