Paper “Decentralized Multi-Agent Multi-Task Q-Learning with Function Approximation for POMDPs” by M.S. Stanković, M. Beko and S.S. Stanković, has been accepted for presentation and publication in the proceedings of the 63rd IEEE Conference on Decision and Control!

The paper proposes a novel distributed gradient-based two-time-scale algorithm for decentralized multi-agent multi-task learning (MTL) using a linear approximation of the optimal action value function (Q-function) in POMDPs. The algorithm is based on the idea of using in a concurrent way recursive Bayesian state belief filters for estimation of the system model parameters, prediction of the hidden state and definition of the optimal approximation parameters of the local Q-functions. The main MTL algorithm is composed of: 1) local parameter updates based on an off-policy gradient-based learning algorithm with target policy belonging to the greedy or Gibbs classes, and 2) a linear stochastic time-varying consensus scheme for parameters shared between the agents in order to achieve the MTL goal. It is proved, under general assumptions, that the parameter estimates generated by the proposed algorithm weakly converge to a bounded invariant set of the corresponding ordinary differential equations (ODE). Simulation results illustrate the effectiveness of the algorithm.

Paper “Optimizing long-short term memory neural networks for electroencephalogram anomaly detection using variable neighborhood search with dynamic strategy change” by Branislav Radomirovic, Nebojsa Bacanin, Luka Jovanovic, Vladimir Simic, Angelina Njegus, Dragan Pamucar, Mario Köppen and Miodrag Zivkovic has been published in journal Complex & Intelligent Systems.

The paper proposes an innovative approach employing time-series classification of EEG data, leveraging long-short-term memory (LSTM) neural networks for the identification of abnormal brain activity, particularly seizures. To enhance the performance of the proposed model, metaheuristic algorithms were employed for optimizing hyperparameter collection. Additionally, a tailored modification of the variable neighborhood search (VNS) is introduced, specifically tailored for this neurodiagnostic application. The effectiveness of this methodology is evaluated using a carefully curated dataset comprising real-world EEG recordings from both healthy individuals and those affected by epilepsy. This software-based approach demonstrates noteworthy results, showcasing its efficacy in anomaly and seizure detection, even when working with relatively modest sample sizes.

Paper “Evaluating the performance of metaheuristic-tuned weight agnostic neural networks for crop yield prediction” by L. Jovanovic, M. Zivkovic, N. Bacanin, M. Dobrojevic, V. Simic, K. Kumar Sadasivuni & E. Babaee Tirkolaee has been published in journal Neural Computing and Applications.

The published study explores crop yield forecasting through weight agnostic neural networks (WANN) optimized by a modified metaheuristic. WANNs offer the potential for lighter networks with shared weights, utilizing a two-layer cooperative framework to optimize network architecture and shared weights. The proposed metaheuristic is tested on real-world crop datasets and benchmarked against state-of-the-art algorithms using standard regression metrics. The model demonstrates significant potential in crop forecasting with lightweight architectures.

Paper “Distributed Multi-Agent Gradient Based Q-Learning with Linear Function Approximation” by M.S. Stanković, M. Beko and S.S. Stanković, has been accepted for presentation and publication in the proceedings of the 2024 European Control Conference!

The paper proposes a novel multi-agent algorithm for distributed gradient-based two-time-scale off-policy learning of linear approximation of the optimal state-action value function (Q-function) in Markov decision processes. The superior convergence properties of the algorithm have been rigorously proved, and the effectiveness of the algorithm was illustrated using simulations.

Paper “Tuning attention based long-short term memory neural networks for Parkinson’s disease detection using modified metaheuristics” by A. Ćuk, T. Bezdan, L. Jovanović, M. Antonijević, M. S. Stanković, V. Simić, M. Živković & N. Bačanin had been published in journal Scientific Reports.

The paper studies the potential of utilizing long-short term memory neural networks (LSTM) with attention mechanisms to detect Parkinson’s disease based on dual-task walking test data. Given that the performance of neural networks is significantly inductance by architecture and training parameter choices, a modified version of the recently introduced crayfish optimization algorithm (COA) was proposed, specifically tailored to the requirements of this investigation. The proposed optimizer is assessed on a publicly accessible real-world clinical gait in Parkinson’s disease dataset, and the results demonstrate its promise, achieving an accuracy of 87.4187 % for the best-constructed models.

Paper “The Serbian Sign Language Alphabet: A Unique Authentic Dataset of Letter Sign Gestures“, by M. Radaković, M. Marjanović, I. Ristić, V. Kuleto, M. P. Ilić, and S. Dabić-Miletić has been published in journal Mathematics!

This article focuses on detailing a Serbian Sign Language alphabet database and the method for creating it in order to provide a foundation for answering the various societal challenges of persons who use the Serbian language. The proposed techniques involve machine learning using optimized convolutional neural network models.

Paper “Optimizing long-short-term memory models via metaheuristics for decomposition aided wind energy generation forecast“, by M. Pavlov-Kagadejev, L. Jovanovic, N. Bacanin, M. Deveci, M. Zivkovic, M. Tuba, I. Strumberger & W. Pedrycz has been published in journal Artificial Intelligence Review!

The paper proposed an approach that incorporates signal decomposition techniques with Long Short-Term Memory (LSTM) neural networks tuned via a modified metaheuristic algorithm used for wind power generation forecasting. LSTM networks perform notably well when addressing time-series prediction, and further hyperparameter tuning by a modified version of the reptile search algorithm (RSA) can help improve performance. The modified RSA was first evaluated against standard CEC2019 benchmark instances before being applied to the practical challenge. The proposed tuned LSTM model has been tested against two wind production datasets with hourly resolutions. The predictions were executed without and with decomposition for one, two, and three steps ahead. Simulation outcomes have been compared to LSTM networks tuned by other cutting-edge metaheuristics. It was observed that the introduced methodology notably exceed other contenders, as was later confirmed by the statistical analysis.