http://popups.lib.uliege.be/1373-5411/index.php?id=146 Publications of Auteurs Tadashi Ae fr 0 http://popups.lib.uliege.be/1373-5411/index.php?id=2937 We focus on sequences of the data of which a user selects from a multimedia database. These data cannot be stereotyped because user's view for them changes by each user. Therefore, we represent the structure of the multimedia database as the vector representing both user's view and the stereotyped vector. Such a vector can be classified by SOM (Self-Organizing Map). On the other hand, we introduce a technique for data set generation. If such a set consists of sequences of data, Hidden Markov Model (HMM) will be available for practical purposes. Therefore, we introduce HMM and Vector-state Markov Model (VMM) to represent the vector of user's view, and to acquire the sequence containing the change of user's view. Lastly, we will refer to an extended technique for an interactive system using the rough set theory. Tue, 03 Sep 2024 15:56:25 +0200 Tue, 08 Oct 2024 14:44:43 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=2937 http://popups.lib.uliege.be/1373-5411/index.php?id=2605 This paper describes an approach to generate a sequence requiring an unrealizable function by programs, such as a flash that is required especially in creative activity of a human. We have already proposed a recurrent neural network that demonstrates a generation of several creative sequences, but convergency and stability problems occur. On the other hand, it is known in biological experiments where the chaotic sequences can be observed from brain waves. The neural network constructed from chaotic neurons has nonlinear dynamics, but there remains the difficulty of training method. We propose an evolutional methodology to train a chaotic neural network, and introduce Darwinism for its evolving process. To determine their most suitable structure and the weights of connection, we use AIC for the fitness value. Thu, 29 Aug 2024 15:07:30 +0200 Tue, 08 Oct 2024 14:04:11 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=2605 http://popups.lib.uliege.be/1373-5411/index.php?id=1626 First, we introduce VM(vector state machine) which is generalized from the structured vector addition system, and, next, propose KR/VM model, where KR is the model for knowledge representation, since the conventional AI (Artificial Intelligence) technique is also important. As a result, we obtain a hybrid model of the AI model and the vector state machine, and it will be a good brain agent model which is widely applicable for the practical problems. Mon, 15 Jul 2024 16:04:47 +0200 Mon, 15 Jul 2024 16:04:55 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=1626 http://popups.lib.uliege.be/1373-5411/index.php?id=1482 We believe the quantum functional device to be a future perspective device, if we solve the problems that it has nowadays. We will summarize such problems with several discussions from the viewpoint of circuit and system. Fri, 12 Jul 2024 15:13:04 +0200 Fri, 12 Jul 2024 15:13:14 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=1482 http://popups.lib.uliege.be/1373-5411/index.php?id=626 A two-level processing scheme for real-time image understanding is proposed, where an example-based (or case-based) reasoning in neural AI systems is introduced. The system has two levels; Component Level and Structure Level. At the component level, an elementary pattern recognition is performed as in the conventional pattern recognition, while the syntax pattern recognition is done at the structure level. Both levels are essentially time-consuming (theoretically, NP-complete each). The pattern recognition assisted by syntax recognition reduces the total complexity of processes, and the system can perform a real-time image understanding, when the VLSI chips are introduced. As a result, we show a reasonable real-time image understanding scheme by introducing neural pattern recognition at the component level and a case-based AI technique at the structure level. Fri, 28 Jun 2024 15:24:01 +0200 Tue, 08 Oct 2024 14:06:02 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=626 http://popups.lib.uliege.be/1373-5411/index.php?id=299 In this paper we introduce an extended vector addition system, i.e., a structured vector addition system. The vector addition system (in short, VAS) is proposed by R. Karp et al. as a parallel processing model, but the VAS seems to be far from the neural network. However it is an excellent "macro" model for the brain behavior, especially, for the emotional behavior. The original VAS is weak to represent the control mechanism, and therefore, we propose a structured VAS (in short, SVAS),where the control mechanism plays a role of simulating the dynamical behavior of human emotion, especially, with the state transition of vectors. We will discuss the inductive learning and the anticipation on SVAS. Fri, 21 Jun 2024 09:45:59 +0200 Fri, 21 Jun 2024 09:46:17 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=299 http://popups.lib.uliege.be/1373-5411/index.php?id=141 A lot of research for anticipatory systems have been reported, where the chaotic equation including the hyper incursion equation plays an important role. The neural network model is also included in such a category and will continue to be discussed. From the viewpoint of computer systems, however, we have proposed a hybrid system architecture mixed with neural network and artificial intelligence, where the two-level structure is introduced ; the first layer : a neural network, and the second layer : an automaton system. On the two-layered system, the automaton part is dominant for anticipation, because the state transition is made by an automaton behavior although the selection among transitions is made by a neural network. In this paper, we discuss an automaton-based anticipation, since it is appropriate to discuss anticipation together with learnability. Tue, 18 Jun 2024 16:19:12 +0200 Mon, 07 Oct 2024 12:53:35 +0200 http://popups.lib.uliege.be/1373-5411/index.php?id=141