Applications of the self-organising map to reinforcement learning

Unsupervised learning of perceptual feature combinations

In many situations it is behaviorally relevant for an animal to respond to co-occurrences of perceptual, possibly polymodal features, while these features alone may have no importance. Thus, it is crucial for animals to learn such feature combinations in spite of the fact that they may occur with variable intensity and occurrence frequency. Here, we present a novel unsupervised learning mechanism that is largely independent of these contingencies and allows neurons in a network to achieve specificity for different feature combinations. This is achieved by a novel correlation-based (Hebbian) learning rule, which allows for linear weight growth and which is combined with a mechanism for gradually reducing the learning rate as soon as the neuron's response becomes feature combination specific. In a set of control experiments, we show that other existing advanced learning rules cannot satisfactorily form ordered multi-feature representations. In addition, we show that networks, which use this type of learning always stabilize and converge to subsets of neurons with different feature-combination specificity. Neurons with this property may, thus, serve as an initial stage for the processing of ecologically relevant real world situations for an animal.

Safe deep reinforcement learning in diesel engine emission control

A deep reinforcement learning application is investigated to control the emissions of a compression ignition diesel engine. The main purpose of this study is to reduce the engine-out nitrogen oxide ( N O x ) emissions and to minimize fuel consumption while tracking a reference engine load. First, a physics-based engine simulation model is developed in GT-Power and calibrated using experimental data. Using this model and a GT-Power/Simulink co-simulation, a deep deterministic policy gradient is developed. To reduce the risk of an unwanted output, a safety filter is added to the deep reinforcement learning. Based on the simulation results, this filter has no effect on the final trained deep reinforcement learning; however, during the training process, it is crucial to enforce constraints on the controller output. The developed safe reinforcement learning is then compared with an iterative learning controller and a deep neural network-based nonlinear model predictive controller. This comparison shows that the safe reinforcement learning is capable of accurately tracking an arbitrary reference input while the iterative learning controller is limited to a repetitive reference. The comparison between the nonlinear model predictive control and reinforcement learning indicates that for this case reinforcement learning is able to learn the optimal control output directly from the experiment without the need for a model. However, to enforce output constraint for safe learning reinforcement learning, a simple model of system is required. In this work, reinforcement learning was able to reduce N O x emissions more than the nonlinear model predictive control; however, it suffered from slightly higher error in load tracking and a higher fuel consumption.

The dilemma of the symbols: analogies between philosophy, biology and artificial life

This article analyzes some analogies going from Artificial Life questions about the symbol-matter connection to Artificial Intelligence questions about symbol-grounding. It focuses on the notion of the interpretability of syntax and how the symbols are integrated in a unity ("binding problem"). Utilizing the DNA code as a model, this paper discusses how syntactic features could be defined as high-grade characteristics of the non syntactic relations in a material-dynamic structure, by using an emergentist approach. This topic furnishes the ground for a confutation of J. Searle's statement that syntax is observer-relative, as he wrote in his book "Mind: A Brief Introduction". Moreover the evolving discussion also modifies the classic symbol-processing doctrine in the mind which Searle attacks as a strong AL argument, that life could be implemented in a computational mode. Lastly, this paper furnishes a new way of support for the autonomous systems thesis in Artificial Life and Artificial Intelligence, using, inter alia, the "adaptive resonance theory" (ART).

Modeling and Simulation of Elementary Robot Behaviors using Associative Memories

Today, there are several drawbacks that impede the necessary and much needed use of robot learning techniques in real applications. First, the time needed to achieve the synthesis of any behavior is prohibitive. Second, the robot behavior during the learning phase is – by definition – bad, it may even be dangerous. Third, except within the lazy learning approach, a new behavior implies a new learning phase. We propose in this paper to use associative memories (self-organizing maps) to encode the non explicit model of the robot-world interaction sampled by the lazy memory, and then generate a robot behavior by means of situations to be achieved, i.e., points on the self-organizing maps. Any behavior can instantaneously be synthesized by the definition of a goal situation. Its performance will be minimal (not necessarily bad) and will improve by the mere repetition of the behavior.

An Ensemble of Cooperative Extended Kohonen Maps for Complex Robot Motion Tasks

Self-organizing feature maps such as extended Kohonen maps (EKMs) have been very successful at learning sensorimotor control for mobile robot tasks. This letter presents a new ensemble approach, cooperative EKMs with indirect mapping, to achieve complex robot motion. An indirect-mapping EKM self-organizes to map from the sensory input space to the motor control space indirectly via a control parameter space. Quantitative evaluation reveals that indirect mapping can provide finer, smoother, and more efficient motion control than does direct mapping by operating in a continuous, rather than discrete, motor control space. It is also shown to outperform basis function neural networks. Furthermore, training its control parameters with recursive least squares enables faster convergence and better performance compared to gradient descent. The cooperation and competition of multiple self-organized EKMs allow a nonholonomic mobile robot to negotiate unforeseen, concave, closely spaced, and dynamic obstacles. Qualitative and quantitative comparisons with neural network ensembles employing weighted sum reveal that our method can achieve more sophisticated motion tasks even though the weighted-sum ensemble approach also operates in continuous motor control space.

Auditory learning: a developmental method

Motivated by the human autonomous development process from infancy to adulthood, we have built a robot that develops its cognitive and behavioral skills through real-time interactions with the environment. We call such a robot a developmental robot. In this paper, we present the theory and the architecture to implement a developmental robot and discuss the related techniques that address an array of challenging technical issues. As an application, experimental results on a real robot, self-organizing, autonomous, incremental learner (SAIL), are presented with emphasis on its audition perception and audition-related action generation. In particular, the SAIL robot conducts the auditory learning from unsegmented and unlabeled speech streams without any prior knowledge about the auditory signals, such as the designated language or the phoneme models. Neither available before learning starts are the actions that the robot is expected to perform. SAIL learns the auditory commands and the desired actions from physical contacts with the environment including the trainers.