AI Works Beyond The Grave

Halloween might be behind us, but that doesn’t mean we have to forget those who have passed into another world. That’s why this week’s newsletter will dip into a ‘recent’ interview with Steve Jobs, alongside photos of Freddie Mercury and John Lennon as they’d look if they were alive today. And if that’s not enough … Read more

Unique, chip-based light source allows scanning lidar to be combined with 3D flash lidar — ScienceDaily

Our roads might one day be safer thanks to a completely new type of system that overcomes some of lidar’s limitations. Lidar, which uses pulsed lasers to map objects and scenes, helps autonomous robots, vehicles and drones to navigate their environment. The new system represents the first time that the capabilities of conventional beam-scanning lidar … Read more

Incremental Machine Learning for Linked Data Event Streams | by Samuel Van Ackere | Feb, 2023

Unlocking the Power of Real-time Predictions: An Introduction to Incremental Machine Learning for Linked Data Event Streams Photo by Isaac Smith on Unsplash This article discusses online machine learning, one of the most exciting subdomains of machine learning theory. The potential of using incremental machine learning becomes more and more apparent when working on fast-moving … Read more

How undesired goals can arise with correct rewards

Exploring examples of goal misgeneralisation – where an AI system’s capabilities generalise but its goal doesn’t As we build increasingly advanced artificial intelligence (AI) systems, we want to make sure they don’t pursue undesired goals. Such behaviour in an AI agent is often the result of specification gaming – exploiting a poor choice of what … Read more

Measuring perception in AI models

New benchmark for evaluating multimodal systems based on real-world video, audio, and text data From the Turing test to ImageNet, benchmarks have played an instrumental role in shaping artificial intelligence (AI) by helping define research goals and allowing researchers to measure progress towards those goals. Incredible breakthroughs in the past 10 years, such as AlexNet … Read more

Incorporating Sentiment Analysis into E-commerce

October 7, 2022 4 min read By Cogito Tech LLC. Studying users’ behavior and understanding their sentiments have become substantial to businesses with the increasing platforms operating in the digital space and users’ traffic rapidly growing on these online platforms. The Internet replaced sources such as friends, relatives, and consumer reports which used to be … Read more

Attaining XGBoost-level performance with the interpretability and speed of CART – The Berkeley Artificial Intelligence Research Blog

FIGS (Fast Interpretable Greedy-tree Sums): A method for building interpretable models by simultaneously growing an ensemble of decision trees in competition with one another.

Recent machine-learning advances have led to increasingly complex predictive models, often at the cost of interpretability. We often need interpretability, particularly in high-stakes applications such as in clinical decision-making; interpretable models help with all kinds of things, such as identifying errors, leveraging domain knowledge, and making speedy predictions.

In this blog post we’ll cover FIGS, a new method for fitting an interpretable model that takes the form of a sum of trees. Real-world experiments and theoretical results show that FIGS can effectively adapt to a wide range of structure in data, achieving state-of-the-art performance in several settings, all without sacrificing interpretability.

Read more

Preparing for the Unexpected

Some of the problems we tackle using machine learning involve categorical features that represent real world objects, such as words, items and categories. So what happens when at inference time we get new object values that have never been seen before? How can we prepare ourselves in advance so we can still make sense out … Read more