Paper “Modified Artificial Bee Colony Algorithm for Tuning Simple LSTM for Multivariate Time-Series Forecasting,” by J. Krstovic, N. Bacanin, M. Zivkovic, A. Bozovic, M.S. Stankovic, M. Antonijevic, and T. Bezdan, has been accepted for presentation at the 5th International Conference on Communication and Computational Technologies (ICCCT 2023) to be held in January 2023.
Paper “Distributed Consensus-Based Multi-Agent Temporal-Difference Learning“, by M.S. Stanković, M. Beko and S.S. Stanković, has been published in Automatica!
The paper proposes two new distributed consensus-based algorithms for temporal-difference learning in multi-agent Markov decision processes. The algorithms are of off-policy type and are aimed at linear approximation of the value function. Restricting agents’ observations to local data and communications to their small neighborhoods, the algorithms consist of: a) local updates of the parameter estimates based on either the standard TD(λ) or the emphatic ETD(λ) algorithm, and b) dynamic consensus scheme implemented over a time-varying lossy communication network. The algorithms are completely decentralized, allowing efficient parallelization and applications where the agents may have different behavior policies and different initial state distributions while evaluating a common target policy. It was proved under nonrestrictive assumptions that the proposed algorithms weakly converge to the solutions of the mean ordinary differential equation (ODE) common for all the agents. It was also proved that the whole system may be stabilized by a proper choice of the network and that the parameter estimates weakly converge to consensus. Simulation results were also presented, illustrating the main properties of the algorithms and providing comparisons with similar existing schemes.
Paper “IR Building Analysis with Extraction of Elements Using Image Segmentation and RetinaNet“, by I. Walter, M. Tanasković, and M.S. Stanković. has been published in journal Buildings!
In December 2022, Jelena Gavrilović, DECIDE project team member, has defended her PhD dissertation titled “Jedna klasa sistema za generisanje i distribuciju kriptoloških ključeva zasnovana na više biometrijskih modaliteta“. Congratulations!
Paper “Analysis of Available Data and Parameters in the Process of Forensics of Unmanned Aerial Vehicles“, by J. Gavrilović, M. Tanasković, M.S. Stanković, and A. Ćuk., has been published in Journal of Mechatronics, Automation and Identification Technology (JMAIT), Vol. 7, No. 1, pp. 8 – 11, 2022.
Paper “Tuning Multi-Layer Perceptron by Hybridized Arithmetic Optimization Algorithm for Healthcare 4.0”, by Marko Stankovic, J. Gavrilovic, D. Jovanovic, M. Zivkovic, M. Antonijevic, N. Bacanin, and M.S. Stankovic, has been presented at the 4th International Conference on Innovative Data Communication Technology and Application (ICIDCA), 2022.
Paper “Multi-Agent Actor-Critic Multitask Reinforcement Learning Based on Consensus,” by Milos S. Stankovic, Marko Beko, Nemanja Ilic and Srdjan S. Stankovic, has been presented at the renowned IEEE Conference on Decision and Control (CDC) held in December 2022 in Cancun, Mexico.
Paper “Tuning Machine Learning Models Using a Group Search Firefly Algorithm for Credit Card Fraud Detection“, by D. Jovanovic, M. Antonijevic, M.S. Stankovic, M. Zivkovic, M. Tanaskovic and N. Bacanin, has been published in Mathematics!
The research published in the paper proposes a hybrid machine learning and swarm metaheuristic approach to address the challenge of credit card fraud detection. The novel, enhanced firefly algorithm, named group search firefly algorithm, was devised and then used to a tune support vector machine, an extreme learning machine, and extreme gradient-boosting machine learning models. Boosted models were tested on the real-world credit card fraud detection dataset. The performance of the proposed group search firefly metaheuristic was compared with other recent state-of-the-art approaches. The experimental findings clearly demonstrate that the models tuned by the proposed algorithm obtained superior results in comparison to other models hybridized with competitor metaheuristics.
Paper “Feature Selection by Hybrid Artificial Bee Colony Algorithm for Intrusion Detection,” by M.S. Stankovic, M. Antonijevic, N. Bacanin, M. Zivkovic, M. Tanaskovic, D. Jovanovic, has been accepted for presentation at International Conference on Edge Computing and Applications ICECAA 2022, to be held in October 2022.
Paper “Convergent Distributed Actor-Critic Algorithm Based on Gradient Temporal Difference”, by M.S. Stanković, M. Beko, and S.S. Stankovic, has been presented at the conference EUSIPCO 2022, Belgrade, Serbia, September 2022.