Modeling pheromone source localization behavior by an insect

2016 Fiscal Year Final Research Report

• Project/Area Number: 25870915
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Soft computing; Intelligent informatics
• Research Institution: National Institute of Informatics
• Principal Investigator: Kobayashi Ryota 国立情報学研究所, 情報学プリンシプル研究系, 助教 (70549237)
• Project Period (FY): 2013-04-01 – 2017-03-31
• Keywords: 知能情報学 / 確率過程 / ソフトコンピューティング / シミュレーション / ベイズ統計学

Outline of Final Research Achievements

Some insects, including ants, bees, and moths, can locate the source of pheromone based on the olfactory information. Especially, male silk moth can reach his mate (pheromone source) from sparse pheromone detections in spite of his weak eyesight. In this study we develop a search algorithm by mimicking the procedure that an insect searched a pheromone source. The pheromone detections of a searcher are described by using a Point process model and an algorithm for estimating the position of the source is derived by applying the framework of Bayesian statistics. We also evaluate the performance of search algorithms by simulating the agent that locates a fixed pheromone source. In addition, we have developed a method for predicting the future access to an online content by applying the framework of Point process.

Free Research Field

計算論的神経科学，機械学習