000104730 001__ 104730
000104730 005__ 20190316233958.0
000104730 0247_ $$2doi$$a10.1109/CEC.2007.4424781
000104730 037__ $$aCONF
000104730 245__ $$aEvolving Neuromodulatory Topologies for Reinforcement Learning-like Problems
000104730 269__ $$a2007
000104730 260__ $$bIEEE Press$$c2007
000104730 336__ $$aConference Papers
000104730 520__ $$aEnvironments with varying reward contingencies constitute a challenge to many living creatures. In such conditions, animals capable of adaptation and learning derive an advantage. Recent studies suggest that neuromodulatory dynamics are a key factor in regulating learning and adaptivity when reward conditions are subject to variability. In biological neural networks, specific circuits generate modulatory signals, particularly in situations that involve learning cues such as a reward or novel stimuli. Modulatory signals are then broadcast and applied onto target synapses to activate or regulate synaptic plasticity. Artificial neural models that include modulatory dynamics could prove their potential in uncertain environments when online learning is required. However, a topology that synthesises and delivers modulatory signals to target synapses must be devised. So far, only handcrafted architectures of such kind have been attempted. Here we show that modulatory topologies can be designed autonomously by artificial evolution and achieve superior learning capabilities than traditional fixed-weight or Hebbian networks. In our experiments, we show that simulated bees autonomously evolved a modulatory network to maximise the reward in a reinforcement learning-like environment.
000104730 6531_ $$aAGE
000104730 6531_ $$aAnalog Genetic Encoding
000104730 6531_ $$aNeuroevolution
000104730 6531_ $$aNeuromodulation
000104730 6531_ $$aReinforcement Learning
000104730 6531_ $$aImplicit Encoding
000104730 6531_ $$aImplicit Genetic Encoding
000104730 6531_ $$aEvolutionary Robotics
000104730 700__ $$aSoltoggio, Andrea
000104730 700__ $$0243223$$aDürr, Peter$$g167254
000104730 700__ $$0241582$$aMattiussi, Claudio$$g140974
000104730 700__ $$0240742$$aFloreano, Dario$$g111729
000104730 7112_ $$aIEEE Congress on Evolutionary Computation (CEC 2007) 25-28 Sept. 2007$$cSingapore$$dSeptember 25-28, 2007
000104730 773__ $$q2471-2478$$tProceedings of the 2007 IEEE Congress on Evolutionary Computation
000104730 8564_ $$s279672$$uhttps://infoscience.epfl.ch/record/104730/files/SoltoggioDuerrMattiussiFloreano2007.pdf$$zn/a
000104730 909C0 $$0252161$$pLIS$$xU10370
000104730 909CO $$ooai:infoscience.epfl.ch:104730$$pconf$$pSTI$$qGLOBAL_SET
000104730 917Z8 $$x255330
000104730 937__ $$aLIS-CONF-2007-003
000104730 973__ $$aEPFL$$rREVIEWED$$sPUBLISHED
000104730 980__ $$aCONF