Table 2: The accuracy on the VQA v2.0 test set. We thank all the reviewers for the helpful comments. Q1: How the paper's contribution relates to the current SOT A? SGAE is a rather complicated scene-graph based method specific to image captioning. The results with current SOT A + MIA will be stated more clearly in the paper. Q2: How to use MIA on the baseline systems (i.e., how is MIA applied to image captioning For the settings, we have listed them in the supplementary materials.

Artifact removal in electroencephalography (EEG) is a longstanding challenge that significantly impacts neuroscientific analysis and brain-computer interface (BCI) performance. Tackling this problem demands advanced algorithms, extensive noisy-clean training data, and thorough evaluation strategies. This study presents the Artifact Removal Transformer (ART), an innovative EEG denoising model employing transformer architecture to adeptly capture the transient millisecond-scale dynamics characteristic of EEG signals. Our approach offers a holistic, end-to-end denoising solution for diverse artifact types in multichannel EEG data. We enhanced the generation of noisy-clean EEG data pairs using an independent component analysis, thus fortifying the training scenarios critical for effective supervised learning. We performed comprehensive validations using a wide range of open datasets from various BCI applications, employing metrics like mean squared error and signal-to-noise ratio, as well as sophisticated techniques such as source localization and EEG component classification. Our evaluations confirm that ART surpasses other deep-learning-based artifact removal methods, setting a new benchmark in EEG signal processing. This advancement not only boosts the accuracy and reliability of artifact removal but also promises to catalyze further innovations in the field, facilitating the study of brain dynamics in naturalistic environments.

People consider creativity to be inherently human. However, artificial intelligence (AI) has reached the stage where it can be creative as well. A recent competition attracted anger from artists after it awarded a prize to an artwork created by an AI model known as Midjourney. And such software is now freely available thanks to the release of a similar model called Stable Diffusion, which is the most efficient of its kind to date. Unions of creative practitioners such as Stop AI Stealing the Show have for some time been raising concerns about the use of AI in creative fields.

Many organizations develop software systems using open source software (OSS), which is risky due to the high possibility of losing support. Contributors are critical for the survival of OSS projects, but very few new contributors remain with OSS projects to become long-time contributors (LTCs). Identification of factors that contribute to become an LTC can help OSS project owners utilize limited resources to retain new contributors. In this paper, we investigate whether we can effectively predict new contributors to OSS repositories becoming long time contributors based on repository and contributor meta-data collected from GitHub. We construct a dataset containing 70,899 observations from 888 most popular repositories with 56,766 contributors.

State-of-the-art models keep changing all the time. As someone who has been doing Kaggle competitions for almost a year now, I find myself coming across a lot of them, doing comparisons, evaluating, and testing them. I thought it would be a good idea to list the best models for each ML task so that you know where to start. Without further ado, let's get started! EfficientNetsV2 outperformed state-of-the-art image classification networks by 2% while training 5–11x times faster which is a huge improvement.

Welcome to the September edition of our best and favorite articles in AI that were published this month. We are a Paris-based company that does Agile data development. This month, we spotted articles about AI surveillance, Deepfake, a documentary from the 60s and much more. Let's kick off with the comic of the month: Let's jump in 1960, we are ten years from HAL 9000 and the first personal computers but people are already thinking about the emergence of Artificial Intelligence. From the late 1950s to the early 1960s, newspapers were full of articles about it.

TLDR; Incorporating a new state of the art machine learning model into a production application is a rewarding yet often frustrating experience. The following post provides tips for production Machine Learning,with examples using the Azure Machine Learning Service. If you are new to Azure you can get a free subscription here. While the tips in the following post, transcend Azure, the Azure Machine Learning Service provides structured tooling for training, deploying, automating, and managing machine learning workflows at production scale. Before writing the first line of AI code ask whether the problem you are solving really needs a state of the art model?

This survey samples from the ever-growing family of adaptive resonance theory (ART) neural network models used to perform the three primary machine learning modalities, namely, unsupervised, supervised and reinforcement learning. It comprises a representative list from classic to modern ART models, thereby painting a general picture of the architectures developed by researchers over the past 30 years. The learning dynamics of these ART models are briefly described, and their distinctive characteristics such as code representation, long-term memory and corresponding geometric interpretation are discussed. Useful engineering properties of ART (speed, configurability, explainability, parallelization and hardware implementation) are examined along with current challenges. Finally, a compilation of online software libraries is provided. It is expected that this overview will be helpful to new and seasoned ART researchers.

Question: can AI vision systems from Microsoft and Google, which are available for free to anybody, identify NSFW (not safe for work, nudity) images? Can this identification be used to automatically censor images by blacking out or blurring NSFW areas of the image? Method: I spent a few hours creating in some rough code in Microsoft office to find files on my computer and send them to Google Vision and Microsoft Vision so they could be analysed. I spent a few hours over the weekend just knocking some very rough code. Yes, they did reasonably well at (a) identifying images that could need censoring and (b) identifying where on the image things should be blocked out. Follow on question: Why aren't sites like Facebook and Instagram automatically deploying this technology to identify images and allowing users to choose whether they wish to see such images? How do we know how much of our internet is already censored invisibly?

Previous work in social network analysis (SNA) has modeled the existence of links from one entity to another, but not the attributes such as language content or topics on those links. We present the Author-Recipient-Topic (ART) model for social network analysis, which learns topic distributions based on the direction-sensitive messages sent between entities. The model builds on Latent Dirichlet Allocation (LDA) and the Author-Topic (AT) model, adding the key attribute that distribution over topics is conditioned distinctly on both the sender and recipient---steering the discovery of topics according to the relationships between people. We give results on both the Enron email corpus and a researcher's email archive, providing evidence not only that clearly relevant topics are discovered, but that the ART model better predicts people's roles and gives lower perplexity on previously unseen messages. We also present the Role-Author-Recipient-Topic (RART) model, an extension to ART that explicitly represents people's roles.