• Progress in Medical Physics
• /
• v.21 no.1
• /
• pp.93-98
• /
• 2010

The aim of this work was to establish the methodology for event positioning by measuring depth of interaction (DOI) information and to evaluate the system sensitivity and spatial resolution of the new detector for I-125 and Tc-99m imaging. For this purpose, a Monte Carlo simulation tool, DETECT2000 and GATE were used to model the energy deposition and light distribution in the detector and to validate this approach. Our proposed detector module consists of a monolithic CsI(Tl) crystal with dimensions of $50.0{\times}50.0{\times}3.0\;mm^3$. The results of simulation demonstrated that the resolution is less than 1.5 mm for both I-125 and Tc-99m. The main advantage of the proposed detector module is that by using 3 mm thick CsI(Tl) with maximum-likelihood position-estimation (MLPE) method, high resolution I-125 imaging and high sensitivity Tc-99m imaging are possible. In this paper, we proved that our new detector to be a reliable design as a detector for a multi-energy SPECT.

• The KIPS Transactions:PartB
• /
• v.9B no.2
• /
• pp.139-146
• /
• 2002

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to choose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input-output pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless these algorithms can learn the optimal policy if the agent can visit every state-action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem, we suggest Adaptive Mediation-based Modular Q-Learning (AMMQL) as an improvement of the existing Modular Q-Learning (MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state space effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. In this paper we use the AMMQL algorithn as a learning method for dynamic positioning of the robot soccer agent, and implement a robot soccer agent system called Cogitoniks.