Deep learning in neural networks: An overview
J Schmidhuber - Neural networks, 2015 - Elsevier
In recent years, deep artificial neural networks (including recurrent ones) have won
numerous contests in pattern recognition and machine learning. This historical survey …
numerous contests in pattern recognition and machine learning. This historical survey …
Neural modularity helps organisms evolve to learn new skills without forgetting old skills
A long-standing goal in artificial intelligence is creating agents that can learn a variety of
different skills for different problems. In the artificial intelligence subfield of neural networks …
different skills for different problems. In the artificial intelligence subfield of neural networks …
Real-time neuroevolution in the NERO video game
KO Stanley, BD Bryant… - IEEE transactions on …, 2005 - ieeexplore.ieee.org
In most modern video games, character behavior is scripted; no matter how many times the
player exploits a weakness, that weakness is never repaired. Yet, if game characters could …
player exploits a weakness, that weakness is never repaired. Yet, if game characters could …
[PDF][PDF] Accelerated Neural Evolution through Cooperatively Coevolved Synapses.
Many complex control problems require sophisticated solutions that are not amenable to
traditional controller design. Not only is it difficult to model real world systems, but often it is …
traditional controller design. Not only is it difficult to model real world systems, but often it is …
Evolutionary computation for reinforcement learning
S Whiteson - Reinforcement Learning: State-of-the-art, 2012 - Springer
Algorithms for evolutionary computation, which simulate the process of natural selection to
solve optimization problems, are an effective tool for discovering high-performing …
solve optimization problems, are an effective tool for discovering high-performing …
A system for robotic heart surgery that learns to tie knots using recurrent neural networks
Tying suture knots is a time-consuming task performed frequently during minimally invasive
surgery (MIS). Automating this task could greatly reduce total surgery time for patients …
surgery (MIS). Automating this task could greatly reduce total surgery time for patients …
Driven by compression progress: A simple principle explains essential aspects of subjective beauty, novelty, surprise, interestingness, attention, curiosity, creativity, art …
J Schmidhuber - Workshop on anticipatory behavior in adaptive learning …, 2008 - Springer
I argue that data becomes temporarily interesting by itself to some self-improving, but
computationally limited, subjective observer once he learns to predict or compress the data …
computationally limited, subjective observer once he learns to predict or compress the data …
On learning to think: Algorithmic information theory for novel combinations of reinforcement learning controllers and recurrent neural world models
J Schmidhuber - arXiv preprint arXiv:1511.09249, 2015 - arxiv.org
This paper addresses the general problem of reinforcement learning (RL) in partially
observable environments. In 2013, our large RL recurrent neural networks (RNNs) learned …
observable environments. In 2013, our large RL recurrent neural networks (RNNs) learned …
JSBSim: An open source flight dynamics model in C++
J Berndt - … and Simulation Technologies Conference and Exhibit, 2004 - arc.aiaa.org
This paper gives an overview of JSBSim, an open source, multi-platform, flight dynamics
model (FDM) framework written in the C++ programming language. JSBSim is designed to …
model (FDM) framework written in the C++ programming language. JSBSim is designed to …
Discovering parametric activation functions
G Bingham, R Miikkulainen - Neural Networks, 2022 - Elsevier
Recent studies have shown that the choice of activation function can significantly affect the
performance of deep learning networks. However, the benefits of novel activation functions …
performance of deep learning networks. However, the benefits of novel activation functions …