Pattern Recognition


TensorFlow adds a new library for on-device text-to-image search

#artificialintelligence

TensorFlow has announced a new on-device embedding-based search library feature that allows one to quickly find similar images, text or audio from millions of data samples in a few milliseconds. It works by using a model to embed the search query into a high-dimensional vector representing the semantic meaning of the query. Then it uses ScaNN (Scalable Nearest Neighbors) to search for similar items from a predefined database. Given below is a walkthrough of an end-to-end example of building a text-to-image search feature (retrieve the images given textual queries) using the new TensorFlow Lite Searcher Library. The dual encoder model consists of an image encoder and a text encoder.


CS229: Machine Learning - AI Summary

#artificialintelligence

CS229: Machine Learning Course Description This course provides a broad introduction to machine learning and statistical pattern recognition. Topics include: supervised learning (generative/discriminative learning, parametric/non-parametric learning, neural networks, support vector machines); unsupervised learning (clustering, dimensionality reduction, kernel methods); learning theory (bias/variance tradeoffs, practical advice); reinforcement learning and adaptive control. The course will also discuss recent applications of machine learning, such as to robotic control, data mining, autonomous navigation, bioinformatics, speech recognition, and text and web data processing. Course Description This course provides a broad introduction to machine learning and statistical pattern recognition. Topics include: supervised learning (generative/discriminative learning, parametric/non-parametric learning, neural networks, support vector machines); unsupervised learning (clustering, dimensionality reduction, kernel methods); learning theory (bias/variance tradeoffs, practical advice); reinforcement learning and adaptive control.


Cats purrfectly know their feline friends' names, Japanese study says

The Japan Times

Your cat may not seem to be listening to you, but it's now believed that they recognize the names of their feline friends, and maybe your name as well, according to a recent study by Japanese researchers. Dogs have been known to follow human speech to some degree, but it had not been scientifically clear whether cats have a grasp of human languages. This could be due to a conflict with your ad-blocking or security software. Please add japantimes.co.jp and piano.io to your list of allowed sites. If this does not resolve the issue or you are unable to add the domains to your allowlist, please see out this support page.



Thumbs up: Zoom is adding this handy gesture-recognition feature, and more

ZDNet

Video-meeting giant Zoom is rolling out a set of new features including its refreshed digital whiteboard, a backstage for events, gesture recognition, and more. Zoom announced the whiteboard overhaul six months ago and is now rolling the Zoom Whiteboard out to users as a collaboration feature that's always on and built into the desktop client, Zoom Meetings, and Zoom Rooms for third-party touch devices as dedicated whiteboards, such as the DTEN D7 and Neat Board. Which video conferencing platform is right for your business? We've gathered details about 10 leading services. Support for Zoom Chat is coming soon, according to Zoom, but there's no news of the proposed Zoom Whiteboard integration with Meta's Oculus Horizon Workrooms, which was supposed to ship in early 2022.


Zoom's desktop apps now respond to raised hands and thumbs-up gestures

Engadget

You no longer need to bring out an iPad or iPhone just to use Zoom's gesture recognition. Zoom has updated its Mac and Windows apps with visual gesture support. Raise your hand or give a thumbs-up and you'll send the appropriate reaction. As you might imagine, this promises more natural interaction in virtual classrooms and meetings than you'd get from clicking buttons. The feature requires the latest version of Zoom as of this writing (5.10.3).


Google's multisearch feature lets you search with text and images at the same time

ZDNet

Google on Thursday began rolling out a new Search feature that will let users search for information using both text and images at the same time. The new multisearch feature is part of Google's ongoing efforts to use AI to "create information experiences that are truly conversational, multimodal and personal," as Google CEO Sundar Pichai said recently. These Chromebook laptops feature low prices and long battery lives. The multisearch feature is embedded in Google Lens, the image recognition tool that's accessible via the Google app. For now, the feature is available in beta for US users searching with text in English.


How to tackle machine learning's MLOps tooling mess

#artificialintelligence

We've been overcomplicating machine learning for years. Sometimes we confuse it with the over-hyped artificial intelligence, talking about replacing humans with robotic reasoning when really ML is about augmenting human intelligence with advanced pattern recognition. Or we burrow into deep learning when more basic SQL queries would get the job done. But perhaps the greatest problem with ML today is how incredibly complicated we make the tooling because, as Confetti AI co-founder Mihail Eric has posited, the ML "tooling landscape with constantly shifting responsibilities and new lines in the sand is especially hardest for newcomers to the field," making it "a pretty rough time to be taking your first steps into MLOps." Normally we look to tooling to make tech easier.


AI reveals link between family history and type 1 diabetes risks - Futurity

#artificialintelligence

You are free to share this article under the Attribution 4.0 International license. A new data-driven approach is offering insight into people with type 1 diabetes, who account for about 5-10% of all diabetes diagnoses. The researchers gathered information through health informatics and applied artificial intelligence (AI) to better understand the disease. In the study, they analyzed publicly available, real-world data from about 16,000 participants enrolled in the T1D Exchange Clinic Registry. By applying a contrast pattern mining algorithm, researchers were able to identify major differences in health outcomes among people living with type 1 diabetes who do or do not have an immediate family history of the disease.


Stock Forecast Based On a Predictive Algorithm

#artificialintelligence

This stock forecast is designed for investors and analysts who need predictions of the best stocks for the whole Pharmaceutical sector (see Pharma Stocks Package). Package Name: Pharma Stocks Forecast Recommended Positions: Long Forecast Length: 7 Days (3/16/22 – 3/23/22) I Know First Average: 25.8% The algorithm correctly predicted 10 out of 10 the suggested trades in the Pharma Stocks Forecast Package for this 7 Days forecast. The prediction with the highest return was SPPI, at 85.74%. YMAB and BBIO also performed well for this time horizon with returns of 40.69% and 26.44%, respectively.