International Journal on Artificial Intelligence Tools

A new meta-algorithm for estimating the conditional average treatment effects is pro-posed in the paper. The basic idea behind the algorithm is to consider a new dataset consisting of feature vectors produced by means of concatenation of examples from control and treatment groups, which are close to each other. Outcomes of new data are defined as the difference between outcomes of the corresponding examples comprising new feature vectors. The second idea is based on the assumption that the number of controls is rather large and the control outcome function is precisely determined. This assumption allows us to augment treatments by generating feature vectors which are closed to available treatments. The outcome regression function constructed on the augmented set of concatenated feature vectors can be viewed as an estimator of the conditional average treatment effects. A simple modification of the Co-learner based on the random subspace method or the feature bagging is also proposed. Various numerical simulation experiments illustrate the proposed algorithm and show its outperformance in comparison with the well-known T-learner and X-learner for several types of the control and treatment outcome functions.

The pieces of information that are being collected from regions of disaster is critical as the rapid deployment of the first responders rely on them. Another critical part of that deployment is the acquisition of different types of information (visual, sounds, and others). Even with that information the rescuing teams still face the difficult task of rescuing humans under debris. Some of the constraints that make this task harder are the wrecked building’s unknown structure, time limitation, the difficulty to collect information under the debris and more. An important issue is the accurate collection of information beneath destroyed structures and the 3D representation of the space and the correct location of the human subject under the debris. This paper deals with the design and the capabilities of a ground bio-inspired micro-robot, called Tzitziki, capable to select visual and audio information beneath destroyed buildings and locate human subjects in areas (like deep underground cavities), the reconstruction of the underground cavities and the synergistic collaboration of drones, micro-robots and human first responders. Illustrative examples are provided proving the concept.

As the dairy products have a short consumption period, the accurate prediction of their demand is very important for the dairy industry. Accordingly, this research specifically addresses the prediction of dairy product demand (DPD). The main contribution of this research is to provide an integrated framework based on statistical tests, time-series prediction and artificial intelligence with the runner-root algorithm (RRA) as a novel meta-heuristic algorithm to obtain the best prediction of DPD in Iran. First, a series of economic and social indicators that seemed to be effective in the demand for dairy products are identified and the ineffective indices are eliminated. Next, the artificial intelligence tools including MLP, ANFIS, and LSTM are implemented and improved with the help of RRA. The designed hybrid methods are implemented by using data from 2013 to 2017 of the Iran diary industry. This novel algorithm is compared to gray wolf optimization, invasive weed optimization, and particle swarm optimization. The results show that the proposed MLP-RRA has the most ability to improve by using meta-heuristic algorithms. The coefficient of determination is 98.19%. Moreover, in each artificial intelligence tools, RRA causes better results than the other tested algorithms. The highly accurate results confirm that the proposed hybrid methods based on the RRA algorithm are able to improve the prediction of demand for various products.

A systematic approach is developed for constructing proper trial solutions to Partial Differential Equations (PDEs) of up to second order, using neural forms that satisfy prescribed initial, boundary and interface conditions. The spatial domain considered is of the rectangular hyper-box type. On each face either Dirichlet or Neumann conditions may apply. Robin conditions may be accommodated as well. Interface conditions that induce discontinuities, have not been treated to date in the relevant neural network literature. As an illustration a common problem of heat conduction through a system of two rods in thermal contact is considered.

Stochastic Gradient Descent (SGD) is perhaps the most frequently used method for large scale training. A common example is training a neural network over a large data set, which amounts to minimizing the corresponding mean squared error (MSE). Since the convergence of SGD is rather slow, acceleration techniques based on the notion of “Mini-Batches” have been developed. All of them however, mimicking SGD, impose diminishing step-sizes as a means to inhibit large variations in the MSE objective.

In this article, we introduce random sets of mini-batches instead of individual mini-batches. We employ an objective function that minimizes the average MSE and its variance over these sets, eliminating so the need for the systematic step size reduction. This approach permits the use of state-of-the-art optimization methods, far more efficient than the gradient descent, and yields a significant performance enhancement.

Within this paper, we present two neural nets for view-independent complex human activity recognition (HAR) from video frames. For our study here, we reduce the number of frames produced by a video sequence given that we can identify activities from a sparsely sampled sequence of body poses, and, at the same time, we are able to reduce the processing complexity and response while hardly affecting the accuracy, precision, and recall. To do so, we use a formal framework to ensure the quality of data collection and data preprocessing. We utilize neural networks for the classification of single and complex body activities. More specifically, we consider the sequence of body poses as a time-series problem given that they can provide state-of-the-art results on challenging recognition tasks with little data engineering. Deep Learning in the form of Convolutional Neural Network (CNN), Long Short-Term Neural Network (LSTM), and a one-dimensional Convolutional Neural Network Long Short-Term Memory model (CNN-LSTM) are used as benchmarks to classify the activity.

Traffic Sign Recognition (TSR) is a crucial component in many automotive applications, such as driver assistance, sign maintenance, and vehicle autonomy. In this paper, we present an efficient approach to training a machine learning-based TSR solution. In our choice of recognition method, we have opted for convolutional neural networks, which have demonstrated best-in-class performance in previous works on TSR. One of the challenges related to training deep neural networks is the requirement for a large amount of training data. To circumvent the tedious process of acquiring and manually labelling real data, we investigate the use of synthetically generated images. Our networks, trained on only synthetic data, are capable of recognising traffic signs in challenging real-world footage. The classification results achieved on the GTSRB benchmark are seen to outperform existing state-of-the-art solutions.

Sentiment analysis is a circumstantial analysis of text, identifying the social sentiment to better understand the source material. The article addresses sentiment analysis of an English-Hindi and English-Bengali code-mixed textual corpus collected from social media. Code-mixing is an amalgamation of multiple languages, which previously mainly was associated with spoken language. However, social media users also deploy it to communicate in ways that tend to be somewhat casual. The coarse nature of social media text poses challenges for many language processing applications. Here, the focus is on the low predictive nature of traditional machine learners when compared to Deep Learning counterparts, including the contextual language representation model BERT (Bidirectional Encoder Representations from Transformers), on the task of extracting user sentiment from code-mixed texts. Three deep learners (a BiLSTM CNN, a Double BiLSTM and an Attention-based model) attained accuracy 20–60% greater than traditional approaches on code-mixed data, and were for comparison also tested on monolingual English data.

Face recognition for a single sample per person is challenging due to the lack of sufficient sample information. However, using generic training set to learn an auxiliary dictionary is an effective way to alleviate this problem. Considering generic training sample of diversity, we proposed an algorithm of auxiliary dictionary of diversity learning (ADDL). We first produced virtual face images by mirror images, square block occlusion and grey transform, and then learned an auxiliary dictionary of diversity using a designed objective function. Considering patch-based method can reduce the influence of variations, we seek extended sparse representation with l₂-minimization for each probe patch. Experimental results in the CMUPIE, Extended Yale B and LFW datasets demonstrate that ADDL performs better than other related algorithms.