14532 research outputs found
Sort by
MIPCE: Generating Multiple Patches Counterfactual-Changing Explanations for Time Series Classification
Get PDFIn the development of AI and deep neural networks (DNNs), a growing concern has emerged regarding not only accuracy, but explainability. The corresponding field of research, known as eXplainable AI (XAI), is important because interpreting the predictions of AI helps users make decisions in critical areas such as medicine. XAI has recently gained popularity particularly for counterfactual explanations from a psychological perspective. However, despite recent progress in XAI, few existing methods focus on explaining time series data. We therefore propose Multiple Patches Counterfactual-changing Explanations (MIPCE) for fully convolutional networks (FCNs), which focuses on subsequences of time series, showing the process of change to the counterfactual. First, MIPCE obtains subsequences from features appearing in the FCN, and divides the time series data into patches. Using GPLVM, it then generates the interpretable process of counterfactual change in each patch. We compared our method with other counterfactual methods in terms of proximity, plausibility, and substitutability. These quantitative results indicate that MIPCE outperforms existing methods. In addition, our user test shows that our explanations are useful in helping users understand the decision-making processes of DNNs.Artificial Neural Networks and Machine Learning – ICANN 2023
32nd International Conference on Artificial Neural Networks, Heraklion, Crete, Greece, September 26–29, 202
Study on scintillation properties of Lu2O3-Al2O3 thick film phosphors prepared by high-speed chemical vapor deposition method
Get PDF横浜国立大学博士（工学
Performance Improvement of Single-Electron Reservoir Computing Circuit with Multiple-Tunnel-Junction Single Electron Oscillator
Get PDFWe designed the single-electron (SE) reservoir computing (RC) circuit using multiple-tunnel-junction SE oscillators (MJSEOs), which was compared to a previous SE RC circuit that utilized standard SEOs. The learning performance of the circuits was evaluated by root-mean-square error (RMSE), and it was found that the SE RC circuit with MJSEOs showed superior performance for waveform prediction than the SE RC circuit with SEOs.2023 Silicon Nanoelectronics Workshop (SNW)11-12 June 202
群ロボットのパトロール性能評価における任務環境表現の影響
Get PDFPatrolling is one of the potential applications of multi-robot systems. In a simulation study for patrol algorithms, the way how the study represents the mission environment is also important, as well as the algorithm's design. This study introduces several existing patrol algorithms and evaluates them in grid maps which are different environments than where they have been evaluated. Simulation studies showed that the operation in grid maps, which are suitable to represent large fields, may demonstrate a different performance than maps with other characteristics. The results quantitatively demonstrate the importance of appropriate environmental representation according to the mission characteristics.第41回日本ロボット学会学術講演会 2023年9月11日-2023年9月14
