Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks
By Andrea Soltoggio, Kenneth O. Stanley, Sebastian Risi. 2017
149 citations recorded by the source corpus.
Biological neural networks are systems of extraordinary computational capabilities shaped by evolution, development, and lifelong learning. The interplay of these elements leads to the emergence of biological intelligence. Inspired by such intricate natural phenomena, Evolved Plastic Artificial Neural Networks (EPANNs) employ simulated evolution in-silico to breed plastic neural networks with the aim to autonomously design and create learning systems. EPANN experiments evolve networks that include both innate properties and the ability to change and learn in response to experiences in different environments and problem domains. EPANNs' aims include autonomously creating learning systems, bootstrapping learning from scratch, recovering performance in unseen conditions, testing the computational advantages of particular neural components, and deriving hypotheses on the emergence of biological learning. Thus, EPANNs may include a large variety of different neuron types and dynamics, network architectures, plasticity rules, and other factors. While EPANNs have seen considerable progress over the last two decades, current scientific and technological advances in artificial neural networks are setting the conditions for radically new approaches and results. Exploiting the increased availability of computational resources and of simulation environments, the often challenging task of hand-designing learning neural networks could be replaced by more autonomous and creative processes. This paper brings together a variety of inspiring ideas that define the field of EPANNs. The main methods and results are reviewed. Finally, new opportunities and possible developments are presented.
Subfields: Artificial Intelligence, Model Architectures, Evolutionary Computation
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Jeff Clune
- Christopher Kanan
- Sebastian Risi
- Kenneth O. Stanley
- Joel Lehman
- Yoshua Bengio
- Tyler L. Hayes
- J. Urgen Schmidhuber
In Model Architectures
- Jeff Clune
- Germán Kruszewski
- Thomas Miconi
- Marco Baroni
- Sebastian Risi
- Adam Liska
- Julian Togelius
- Nick Cheney
In Evolutionary Computation
- Jeff Clune
- Sebastian Risi
- Kenneth O. Stanley
- Julian Togelius
- Joel Lehman
- Thomas Miconi
- Adam Liska
- Marco Baroni
Citations in the corpus
- Measuring Catastrophic Forgetting in Neural Networks (2017, background): “Soltoggio, Stanley, and Risi (2017) reviewed neural networks that can adapt their plasticity over time, which they called Evolved Plastic Artificial Neural Networks.”
- Continual Learning with Fully Probabilistic Models (2021, background): “The field of CL is expanding rapidly, see [1, 2, 3, 4] for reviews.”
- Learning to Continually Learn (2020, background): “There is prior work in machine learning that harnesses neuromodulation techniques [7, 16, 39, 40, 42], but in such work neuromodulation directly modulates learning rates, instead of the approach taken in this work of directly modulating activations and thus indirectly controlling learning.”
- AI-GAs: AI-generating algorithms, an alternate paradigm for producing general artificial intelligence (2019, background): “Additionally, there is a recent increase in meta-learning the learning algorithms themselves [37, 191, 40, 187, 161] (discussed more in Section 2.”
- An Investigation of Replay-based Approaches for Continual Learning (2021, background): “Currently, the field is expanding quickly, see [1], [6]–[8] for”
- Backpropamine: training self-modifying neural networks with differentiable neuromodulated plasticity (2018, methodology): “Evolved networks with neuromodulated plasticity were shown to outperform both non-neuromodulated and non-plastic networks in various tasks (e.g. Soltoggio et al. 2008; Risi & Stanley 2012; see Soltoggio et al. 2017 for a review).”
- Meta-learnt priors slow down catastrophic forgetting in neural networks (2019, background): “, (Ellefsen et al., 2015; Soltoggio et al., 2018)), or it can be explicitly optimized to learn good priors (in this case, initial model parameters) that help sequential task learning.”
- Memorize or generalize? Searching for a compositional RNN in a haystack (2018, methodology): “ of neural network plasticity, and we would like to explore architectures that grow larger during training, as they might encourage more modular structures that in turn should favour compositionality [24]. At the same time, with the huge power afforded by these methods come more difficulties in maki”
- Increasing generality in machine learning through procedural content generation (2019, background): “We identify five main open challenges that we believe are essential in pushing the field of PCG forward and to realizing its promise to create more adaptive and lifelong learning ML agents.”