A team of Google researchers recently detailed a framework called SimCLR, which improves previous approaches to self-supervised learning, a family of techniques for converting an unsupervised learning problem (i.e., a problem in which AI models train on unlabeled data) into a supervised one by creating labels from unlabeled data sets. In a preprint paper and accompanying blog post, they say that SimCLR achieved a new record for image classification with a limited amount of annotated data and that it’s simple enough to be incorporated into existing supervised learning pipelines.

That could spell good news for enterprises applying computer vision to domains with limited labeled data.

SimCLR learns basic image representations on an unlabeled corpus and can be fine-tuned with a small set of labeled images for a classification task. The representations are learned through a method called contrastive learning, where the model simultaneously maximizes agreement between differently transformed views of the same image and minimizes agreement between transformed views of different images.

SimCLR first randomly draws examples from the original data set, transforming each sample twice by cropping, color-distorting, and blurring them to create two sets of corresponding views. It then computes the image representation using a machine learning model, after which it generates a projection of the image representation using a module that maximizes SimCLR’s ability to identify different transformations of the same image. Finally, following the pretraining stage, SimCLR’s output can be used as the representation of an image or tailored with labeled images to achieve good performance for specific tasks.

Google says that in experiments SimCLR achieved 85.8% top 5 accuracy on a test data set (ImageNet) when fine-tuned on only 1% of the labels, compared with the previous best approach’s 77.9%.

“[Our results show that] preretraining on large unlabeled image data sets has the potential to improve performance on computer vision tasks,” wrote research scientist Ting Chen and Google Research VP and engineering fellow and Turing Award winner Geoffrey Hinton in a blog post. “Despite its simplicity, SimCLR greatly advances the state of the art in self-supervised and semi-supervised learning.”

Both the code and pretrained models of SimCLR are available on GitHub.

The post Google releases SimCLR, an AI framework that can classify images with limited labeled data appeared first on Artificial Intelligence.

On Thursday, Business Insider reported that Google’s Cloud Vision API service, an AI-powered tool that developers use to identify components in an image like faces, objects, or landmarks, will no longer attach gender-related labels to pictured people.

Yesterday, Google sent out an email to its Cloud Vision API customers that the tool, which can identify and tag various components in an image like brand logos, faces, and landmarks, will no longer attach gender labels like “man” or “woman” to people pictured in an image.

According to the email, as reported by Business Insider, Google said that this practice was to be discontinued because “you can’t deduce someone’s gender by their appearance alone” and doing so would enforce an unethical use of AI. Instead, an individual’s will simply be tagged as a “person”.

Speaking with Business Insider, AI bias expert Frederike Kaltheuner describes this change as “very positive,” stating that “Classifying people as male or female assumes that gender is binary. Anyone who doesn’t fit it will automatically be misclassified and misgendered. So this is about more than just bias — a person’s gender cannot be inferred by appearance. Any AI system that tried to do that will inevitably misgender people.”

Google noted in the email that they intend to continue evolving their AI to ensure that people are not discriminated against based on gender, but also not discriminated against based on factors like, race, ethnicity, income, or religious belief.

The post Google’s AI Tool Will No Longer Attach Gender Labels to People in Pictures appeared first on Artificial Intelligence.