No. 15-16/2022

Online archive of Computer Science and Mathematical Modelling

No. 15-16/2022

1. Radosław Kopiński, Karol Antczak - Image Caption Generation Using Transfer Learning

   Pages: 7 - 12
   Abstract: This paper describes an image caption generation system using deep neural networks. The model is trained to maximize the probability of generated sentence, given the image. The model utilizes transfer learning in the form of pretrained convolutional neural networks to preprocess the image data. The datasets are composed of a still photographs and associated with it, five captions in English language. Constructed model is compared to other similarly constructed models using BLEU score system and ways to further improve its performance are proposed.
   Keywords neural networks, NLP, caption generation, machine learning, computer vision, deep learning, transfer learning.
   Full article: 7_12_rk_ka_image_csmm_15_16_2022.pdf
2. Robert Jarosz - Overview of selected reinforcement learning solutions to several game theory problems

   Pages: 13 - 22
   Abstract: This paper collects several applications of reinforcement learning in solving some problems related to game theory. The methods were selected to possibly show variety of problems and approaches. Selections includes Thompson Sampling, Q-learning, DQN and AlphaGo Zero using Monte Carlo Tree Search algorithm. Paper attempts to show intuition behind proposed algorithms with shallow explaining of technical details. This approach aims at presenting overview of the topic without assuming deep knowledge about statistics and artificial intelligence.
   Keywords artificial intelligence, game theory, Thompson sampling, Q-learning, DQN, Monte Carlo tree search, AlphaZero
   Full article: 13_22_rj_overview_csmm_15_16_2022.pdf
3. Robert Jarosz - Analysis of selected reinforcement learning applications in contract bridge

   Pages: 23 - 31
   Abstract: This paper presents an overview of four selected solutions addressing problem of bidding in card game of contract bridge. In the beginning the basic rules are presented along with basic problem size estimation. Brief description of collected work is presented in chronological order, tracking evolution of approaches to the problem. While presenting solution a short description of mathematical base is attached. In the end a comparison of solution is made, followed by an attempt to estimate future development of techniques.
   Keywords artificial intelligence, bidding, bid prediction, contract bridge, game theory, incomplete knowledge, machine learning, neural networks, Q-learning, reinforcement learning, supervised learning
   Full article: 23_31_rj_analysis_csmm_15_16_2022.pdf
4. Małgorzata Michniewicz - Conclusions from the Analysis of Blockchain Solutions

   Pages: 33 - 37
   Abstract: The development of blockchain technology and distributed ledgers indeed contributes to enabling the delivery of various digital services, such as financial services, registry management, and tokens, including non-fungible tokens (NFTs). In blockchain systems, decisions are made based on the so-called consensus mechanism, which is a method of selecting a single version of transaction history that all nodes consistently agree upon. Although this technology is classified as emerging, it has a history of over ten years, and its ongoing implementations, including in the public administration sector, demonstrate its evolution and its status as one of the most promising technologies in terms of ensuring data immutability. The published ISO (International Organization for Standardization) standards and numerous implementations using business models executed in the DAO (Decentralized Autonomous Organization) architecture undoubtedly support the development of blockchain.
   Keywords blockchain, DAO, public sector
   Full article: 33_37_mm_conclusion_csmm_15_16_2022.pdf
5. Michał Zimoń, Rafał Kasprzyk - When AI fails to see: The challenge of adversarial patches

   Pages: 39 - 44
   Abstract: Object detection, a key application of machine learning in image processing, has achieved significant success thanks to advances in deep learning [6]. In this paper, we focus on analysing the vulnerability of one of the leading object detection models, YOLOv5x [14], to adversarial attacks using specially designed interference known as “adversarial patches” [4]. These disturbances, while often visible, have the ability to confuse the model, which can have serious consequences in real world applications. We present a methodology for generating these interferences using various techniques and algorithms, and we analyse their effectiveness in various conditions. In addition, we discuss potential defences against these types of attacks and emphasise the importance of security research in the context of the growing popularity of ML technology [13]. Our results indicate the need for further research in this area, bearing in mind the evolution of adversarial attacks and their impact on the future of ML technology.
   Keywords object detection, adversarial patches, YOLO model, machine learning
   Full article: 39_44_mz_rk_csmm_15_16_2022.pdf
6. Andrzej Ameljańczyk, Tomasz Ameljańczyk - Fuzzy sets in modelling patientʼs disease states in medical diagnostic support algorithms

   Pages: 45 - 51
   Abstract: The article presents the concept of using fuzzy sets methodology in modelling patientʼs disease states for preliminary medical diagnosis. The preliminary medical diagnosis is based on the identified disease symptoms. The basis of the algorithm are descriptions of the patientʼs disease status and patterns of disease entities. These patterns were defined as fuzzy sets. The paper presents simple classifiers that allow he a preliminary diagnosis based on the analysis of fuzzy sets for the use of the general practitioner.
   Keywords fuzzy set, disease pattern, classifier, preliminary diagnosis.
   Full article: 45_51_aa_ta_fuzzy_csmm_15_16.pdf