…n the 7th AI CITY CHALLENGE (https://www.aicitychallenge.org/), with an absolute advantage of 6% over the runner-up. In this article, we will introduce the competition, our solution and efforts during this event, and a final review. Figure …

…rfit and fail to achieve satisfactory performance on small datasets. To address this issue, data augmentation can be used to generate more data labels or data samples. Data augmentation techniques have been widely used to reduce overfitting…

…eloped an AI technology called Multi-Rate VAE (MR-VAE), enabling the acquisition of the full rate-distortion curve by only a single training run. We will present our work at ICLR 2023 (notable-top-5%). In this blog post, we introduce our jo…

…eloped an AI technology called cost-sensitive self-training that can optimize practical real-world metrics which are complex and present it at NeurIPS 2022. In this blog post, we introduce our joint research. Paper Title: Cost-Sensitive Sel…

…pment on "AI-based Knowledge Discovery". Recently, We have started a joint research project with [Duke University's Interpretable Machine Learning Lab (https://users.cs.duke.edu/~cynthia/lab.html), and the research result, "Exploring the Wh…

…HPCG, HPL-AI, and Graph500), for the fourth consecutive term. At the same conference, we also won the number one spot in the MLPerfTM HPC benchmark, which is dedicated to the actual deep learning (DL) training process. In this blog, we disc…

… Research aims to create “AI that can learn autonomously”, and has been running a joint research program towards this goal with researchers at Massachusetts Institute of Technology (MIT) and Center for Brains, Minds and Machines (CBMM) sinc…

…d like to aim for further popularization of scikit-learn and the acquisition of new contributors from Japan and Asia in general. Please look forward to it! Workshop (Preining) With interpretability of machine learning models ever increasing…

…HPCG, HPL-AI, Graph500), so it is off to a very promising start. My department is involved in researching and developing techniques to accelerate deep learning (DL) processes on Fugaku and PRIMEHPC FX1000/700, which is our product that uses…

We have started a technology blog. I would like to introduce the technologies we are developing as part of our work, report on participation in conferences and events, and provide a variety of other information through this blog. Acknowledg…