動作認識データセット横断の動作ラベル間関係データセット「MetaVD」を公開しました A meta dataset that annotates relationships between action labels across action recognition datasets, MetaVD, is published

action recognition, dataset, MetaVD, NEDO

脆弱性自動検知に向けたバイナリプログラム解析ツールの開発（SCIS2022） Development of a Binary Program Analysis Tool for Automatic Vulnerability Detection

vulnerability, binary analysis, reverse engineering, IoT security

【終了】第3回ステアラボソフトウェア技術セミナー 上里友弥 様「プログラムの最適化手法を用いた Erasure Coding の最適化」 The 3rd STAIR Lab ST Seminar: Dr. Yuya Uezato, "Optimizing Erasure Coding using Program Optimization Techniques"

材料科学論文の図表から情報抽出するツールの開発 Extracting Data From Figures in Materials Science Articles

materials informatics, NEDO, OCR, マテリアルズ・インフォマティクス

Neural generators of sparse local linear models for achieving both accuracy and interpretability Neural generators of sparse local linear models for achieving both accuracy and interpretability

深層学習, deep learning, XAI, ML, 説明可能, ディープラーニング, neural networks

Gaussian Process Regression With Interpretable Sample-wise Feature Weights Gaussian Process Regression With Interpretable Sample-wise Feature Weights

ML, 説明可能, ガウス過程, Gaussian processes, XAI

自己説明型機械学習 Self-Explaining Machine Learning

AI, XAI, 機械学習, ML

【終了】第28回ステアラボ人工知能セミナー 菅原朔 先生「説明性の高い自然言語理解のベンチマークに向けて」 The 28th STAIR Lab AI Seminar: Dr. Saku Sugawara, "Towards a Highly Explainable Benchmark for Natural Language Understanding"

MetaVD: A Meta Video Dataset for enhancing human action recognition datasets MetaVD: A Meta Video Dataset for enhancing human action recognition datasets

action recognition, dataset, STAIR Actions, MetaVD

Predicting How CNN Training Time Changes on Various Mini-Batch Sizes by Considering Convolution Algorithms and Non-GPU Time Predicting How CNN Training Time Changes on Various Mini-Batch Sizes by Considering Convolution Algorithms and Non-GPU Time

GPU performance, Convolutional Neural Networks