A Sequential Learning Resource Allocation Network for Image Processing Applications

Bayesian spiking neurons (BSNs) provide a probablisitic and intuitive interpretation of how spiking neurons could work and have been shown to be equivalent to leaky integrate-and-fire neurons under certain conditions [1]. The study of BSNs has been restricted mainly to small networks because online learning, which currently involves a maximum-likelihood-expectation-maximisation (ML-EM) approach [2,3], is quite slow. Here a new approach to estimating the parameters of Bayesian spiking neurons, referred to as fast learning (FL), is presented and compared to online ML-EM learning. Learning in a BSN is local to the neuron and involves estimation of the transition rate and observation rate parameters of an underlying implicit hidden Markov model (HMM), the hidden state of which the BSN output encodes [1]. Rather than estimating the parameters by maximizing the log-likelihood of the hidden states and the synaptic observations given the parameters as is done in ML-EM [2,3], the FL algorithm directly calculates statistics upon which the parameters depend. This is achieved by taking advantage of the relationship between the log-odds ratio of the hidden state computed by the BSN and the probability that the hidden state is ‘on’ given the past synaptic observations, . Online learning in a two BSN neuron hierarchy is explored, where the first neuron receives N=20 synapses driven by Poisson processes and the second neuron receives input from only the first neuron. Simulations were performed for a fixed set of transition rates and observation rates for 10 different perturbations of the initial transition and observation rate estimates: ±0-20%, ±20-40%,..., ±180-200%. Initial rates were not allowed to go below 10-6. Each perturbation condition was simulated 100 times by randomly selecting the initial parameter values. Although the FL algorithm is not as exact as the ML-EM at estimating the true parameter values for small perturbations of the initial rate estimates relative to the true rates, the FL algorithm is able to reliably estimate the parameters for initial perturbations of up to 200%, whereas the ML-EM algorithm estimates begin to deviate after perturbations of about 40-60%. Moreover, the simplicity of the FL algorithm means that it runs on the order of 25 times faster than the ML-EM implementation considered. These results hold true for both the first and second neurons in the two BSN neuron hierarchy. For the first neuron in the hierarchy, the RMS difference in the time series of the probability calculated for the estimated and the true parameter values, follows a similar pattern to the parameter estimates when the FL and ML-EM algorithms are compared, with average RMS errors of 0.2% obtained for the FL algorithm across the range of perturbations studied. Although we do not have a formal proof of convergence of the FL algorithm, we conclude that the FL algorithm can stably estimate the parameters over a large range of initial perturbations and it can do this very quickly. Thus the FL algorithm makes online learning in networks of BSNs much more tractable.

In this paper, we develop an online sequential learning algorithm for single hidden layer feedforward networks (SLFNs) with additive or radial basis function (RBF) hidden nodes in a unified framework. The algorithm is referred to as online sequential extreme learning machine (OS-ELM) and can learn data one-by-one or chunk-by-chunk (a block of data) with fixed or varying chunk size. The activation functions for additive nodes in OS-ELM can be any bounded nonconstant piecewise continuous functions and the activation functions for RBF nodes can be any integrable piecewise continuous functions. In OS-ELM, the parameters of hidden nodes (the input weights and biases of additive nodes or the centers and impact factors of RBF nodes) are randomly selected and the output weights are analytically determined based on the sequentially arriving data. The algorithm uses the ideas of ELM of Huang et al. developed for batch learning which has been shown to be extremely fast with generalization performance better than other batch training methods. Apart from selecting the number of hidden nodes, no other control parameters have to be manually chosen. Detailed performance comparison of OS-ELM is done with other popular sequential learning algorithms on benchmark problems drawn from the regression, classification and time series prediction areas. The results show that the OS-ELM is faster than the other sequential algorithms and produces better generalization performance.

Bayesian spiking neurons (BSNs) provide a probabilistic interpretation of how neurons perform inference and learning. Online learning in BSNs typically involves parameter estimation based on maximum-likelihood expectation-maximization (ML-EM) which is computationally slow and limits the potential of studying networks of BSNs. An online learning algorithm, fast learning (FL), is presented that is more computationally efficient than the benchmark ML-EM for a fixed number of time steps as the number of inputs to a BSN increases (e.g., 16.5times faster run times for 20 inputs). Although ML-EM appears to converge 2.0 to 3.6times faster than FL, the computational cost of ML-EM means that ML-EM takes longer to simulate to convergence than FL. FL also provides reasonable convergence performance that is robust to initialization of parameter estimates that are far from the true parameter values. However, parameter estimation depends on the range of true parameter values. Nevertheless, for a physiologically meaningful range of parameter values, FL gives very good average estimation accuracy, despite its approximate nature. The FL algorithm therefore provides an efficient tool, complementary to ML-EM, for exploring BSN networks in more detail in order to better understand their biological relevance. Moreover, the simplicity of the FL algorithm means it can be easily implemented in neuromorphic VLSI such that one can take advantage of the energy-efficient spike coding of BSNs.

In this paper, a new classification method based on relevance vector machine (RVM) is used in the MPSK signals classification. Compared with the support vector machine (SVM), RVM is sparse model in the Bayesian framework, not only the solution is highly sparse, but also it does not need to adjust model parameter and its kernel functions don't need to satisfy Mercer's condition. The fourth order cumulants of the received signals are used as the classification vector firstly, and then multi-class classifier of RVM is designed. We first introduce the sparse Bayesian classification model, then transform the RVM learning to the maximization of marginal likelihood, and select the fast sequential sparse Bayesian learning algorithm. Through the experiment compared with SVM classifier proves the advantage of RVM.

An agent must learn and adapt quickly when playing against other agents. This process is challenging in particular when playing in stochastic environments against other learning agents. In this paper, we introduce a fast and adaptive learning algorithm for repeated stochastic games (FAL-SG). FAL-SG utilizes lossy game abstraction to reduce the state space of the game and facilitate learning and adapting rapidly. We analyze FAL-SG's performance by proving bounds on the abstraction loss and prediction mistakes and show that FAL-SG satisfies three criteria prescribed for multiagent learning algorithms. We successfully establish the robustness and scalability of FAL-SG with extensive theoretical and experimental results.

A new fuzzy filter, called fuzzy stack filter (FSF), is proposed to extend the filtering capability of conventional stack filter (SF), which is based on the positive Boolean function (PBF) as its window operator. We fuzzify the onset and off-set of a given PBF to obtain two types of fuzzy PBFs. Then, we adopt the architecture of threshold decomposition to develop this new fuzzy filter with a fuzzy PBF as its window operator. Each fuzzy PBF is associated with a set of control parameters. Therefore, the original PBF can be estimated from above and below by two fuzzy PBFs with appropriate control parameters. Furthermore, we can apply the fuzzy modifiers to modify the fuzzy PBFs such that the PBFs can be completely estimated by the fuzzy PBFs. Hence, the stack filter is a special case of fuzzy stack filter. Since some control parameters are added in this new filter, the neural learning algorithms can be easily developed under the flexibility of the given control parameters. We first propose the fuzzy (m,n) rank-order filter to test our proposed learning algorithm. In this simple learning algorithm, we can remove the noise-corrupted images very well in contrast to the filtering behavior of rank-order filters. We believe that the results presented will lead to more fruitful research on more advanced and powerful learning algorithms dedicated to the appropriate applications.

A fast learning algorithm is proposed for training of multilayer feedforward neural networks, based on a combination of optimal linear Kalman filtering theory and error propagation. In this algorithm, all the information available from the start of the training process to the current training sample is exploited in the update procedure for the weight vector of each neuron in the network in an efficient parallel recursive method. This innovation is a massively parallel implementation and has better convergence properties than the conventional backpropagation learning technique. Its performance is illustrated on some examples, such as a XOR logical operation and a nonlinear mapping of two continuous signals. >

Reinforcement learning is an unsupervised learning algorithm, where learning is based upon feedback from the environment. Prior research has proposed cognitive (e.g., Instance-based Learning or IBL) and statistical (Q-learning) reinforcement learning algorithms. However, an evaluation of these algorithms in a single dynamic environment has not been explored. In this paper, a comparison between the statistical Q-learning algorithm and the cognitive IBL algorithm is presented. A well-known environment, “Frozen Lake,” is used to train, generalize, and scale Q-learning and IBL algorithms. For generalizing, the Q-learning and IBL agents were trained on one version of the Frozen Lake and tested on a permuted version of the same environment. For scaling, the two algorithms were tested on a larger version of the Frozen Lake environment. Results revealed that the IBL algorithm used less training time and generalized better to different environment variants. The IBL algorithm was also able to show scalability by retaining its superior performance in the larger environment. These results indicate that the IBL algorithm could be proposed as an alternative to the standard reinforcement learning algorithms based on dynamic programming such as Q-learning. The inclusion of human factors (such as memory) in the IBL algorithm makes it suitable for robust learning in complex and dynamic environments.

Online sequential extreme learning machine with forgetting mechanism

The paper proposes a fast online learning method for neural network structures by using genetic algorithm (GA) and dynamic back propagation algorithm (BP) jointly. GA is used in the coarse tuning process which adjusts interconnection weights of the neural network. The dynamic back propagation algorithm is subsequently applied to achieve fine adjusting of the network weights. The fitness function, based on the squared error between the teaching signal and the network output value, is redefined at every time step and the proposed GA based algorithm solves a nonstationary function optimization task. At every time step the solution with the best fitness function is used for current representation of the neural network weights and biases. It is shown through the simulations and real time temperature control of drying oven that this learning algorithm has faster convergence ability and better performance on reducing mapping error in the online learning neural network structures. This leads to an improvement of the transient response of neuro adaptive systems. The proposed method has the potential to be applied to many practical areas such as system modeling and control, signal processing and pattern recognition.

The empty nest syndrome has become a significant problem in many countries. Caregivers can help the elderly to cope with some of the issues related to this syndrome, but a tremendous shortage of caregivers is expected due to the low income and high workload. The growing popularity of mobile device use by the elderly has created a great potential for mobile virtual companions. Studies have shown that virtual companions that can show empathy towards its users are perceived as more caring, likeable, trustworthy and supportive. Therefore, we believe that an empathetic companion can help to improve the elderly’s emotional wellness and promote an active independent lifestyle. In this work, we propose an empathetic elderly companion based on the psychological theory of empathy in the form of a smart chat bot. The empathetic companion records the elderly’s speech during their conversations, extracts phonetic features, and employs a fast online learning algorithm, namely Online Sequential Extreme Learning Machine (OS-ELM), to recognize emotion. Based on the elderly’s current emotion, the companion will provide context-aware empathetic responses. The experimental results show that OS-ELM is superior in classifying emotions from phonetic information.

We introduce a new algorithm for sequential learning of Mealy automata by congruence generator extension (CGE). Our approach makes use of techniques from term rewriting theory and universal algebra for compactly representing and manipulating automata using finite congruence generator sets represented as string rewriting systems (SRS). We prove that the CGE algorithm correctly learns in the limit.

Sequential dictionary learning algorithms have been successfully applied to a number of image processing problems. In a number of these problems however, the data used for dictionary learning are structured matrices with notions of smoothness in the column direction. This prior information which can be traduced as a smoothness constraint on the learned dictionary atoms has not been included in existing dictionary learning algorithms. In this paper, we remedy to this situation by proposing a regularized sequential dictionary learning algorithm. The proposed algorithm differs from the existing ones in their dictionary update stage. The proposed algorithm generates smooth dictionary atoms via the solution of a regularized rank-one matrix approximation problem where regularization is introduced via penalization in the dictionary update stage. Experimental results on synthetic and real data illustrating the performance of the proposed algorithm are provided.

Automated decision making based on big data and machine learning (ML) algorithms can result in discriminatory decisions against certain protected groups defined upon personal data like gender, race, sexual orientation etc. Such algorithms designed to discover patterns in big data might not only pick up any encoded societal biases in the training data, but even worse, they might reinforce such biases resulting in more severe discrimination. The majority of thus far proposed fairness-aware machine learning approaches focus solely on the pre-, in- or post-processing steps of the machine learning process, that is, input data, learning algorithms or derived models, respectively. However, the fairness problem cannot be isolated to a single step of the ML process. Rather, discrimination is often a result of complex interactions between big data and algorithms, and therefore, a more holistic approach is required.The proposed FAE (Fairness-Aware Ensemble) framework combines fairness-related interventions at both pre-and post-processing steps of the data analysis process. In the pre-processing step, we tackle the problems of under-representation of the protected group (group imbalance) and of class-imbalance by generating balanced training samples. In the post-processing step, we tackle the problem of class overlapping by shifting the decision boundary in the direction of fairness.

A fast and robust algorithm for training feedforward neural networks (FNNs) by using a variable forgetting factor and U-D factorization-based recursive prediction error (RPE) method is proposed. In comparison with the backpropagation (BP) and RPE based learning algorithms, the proposed algorithm, called UD-RPE, can provide much more accurate learning results in fewer iterations with fewer hidden nodes and improve convergence rate and numerical stability (robustness). In addition, it is less sensitive to start-up parameters, such as initial weights and initial covariance matrix, and the randomness in the observed data. It also has good generalization ability and needs less learning time. Simulation results of nonlinear dynamic system modeling and identification show that the algorithm proposed here is an effective and efficient learning algorithm for FNNs.