At AAAI, our team presented two new multilingual research techniques that enable AI to understand different languages while only trained on one.
PAGs play a vital role in the manufacturing of computer chips. They are also one of several classes of chemical compounds that have recently come under enhanced scrutiny from environmental regulators. Researchers have been racing to create more sustainable ones – but the traditional process of discovering new materials is too slow, too costly, and too risky. So IBM researchers have turned to AI for help – and created new PAGs much, much faster, paving the way to the era of Accelerated Discovery.
Our team has developed an AI that verifies other AIs’ ‘fairness’ by generating a set of counterfactual text samples and testing machine learning systems without supervision.
In a recent paper introduced at the 2021 AAAI Conference on Artificial Intelligence (AAAI), we describe an AI that trades off ‘exploration’ of the world with ‘exploitation’ of its action strategy to maximize rewards. In Reinforcement Learning, an AI gets a reward – such as a bag of gold behind a locked door in a video game – every time it reaches specific desirable states. We have greatly improved this exploration vs exploitation tradeoff using additional commonsense knowledge – in the form of crowdsourced text. Our work could lead to better mapping and navigation applications, and to a new generation of interactive assistive agents able to reason like humans.
We use AI to automatically break down the overall application by representing application code as graphs. Our AI relies on Graph Representation Learning – a popular method in deep learning. Graphs are a natural representation for software and applications. We translated the application to a graph where the programs become nodes. Their relationships with other programs become edges and determine the boundary to separate the nodes of common business functionality.
Our team of researchers from IBM Haifa and Dublin has developed software to help assess privacy risk of AI as well as reduce the amount of personal data in AI training. This software could be of use for fintech, healthcare, insurance, security – or any other industry relying on sensitive data for training.
In our latest paper published in the Microbiome Journal, we propose a way to improve the speed, sensitivity and accuracy of what’s known as microbial functional profiling – determining what microbes in a specific environment are capable of.
Together with Boston Scientific, we are presenting research that details the feasibility and progress towards our new pain measurement method at the 2021 North American Neuromodulation Society Annual Meeting.
Between 2000 and 2001, IBM Research made headlines when it launched an internet-enabled designer watch running Linux, an open-source operating system. Dubbed WatchPad, its aim was to demonstrate the capabilities of the then-novel OS for mobile and embedded devices.
Our team has developed Physics-informed Neural Networks (PINN) models where physics is integrated into the neural network’s learning process – dramatically boosting the AI’s ability to produce accurate results. Described in our recent paper, PINN models are made to respect physics laws that force boundaries on the results and generate a realistic output.
To address the problem of ordinal impacts, our team at IBM T. J. Watson Research Center has developed OGEMs – or Ordinal Graphical Event Models – new dynamic, probabilistic graphical models for events. These models are part of the broader family of statistical and causal models called graphical event models (GEMs) that represent temporal relations where the dynamics are governed by a multivariate point process.
In our recent work, we detail an AI and machine learning mechanism able to assist in correlating a large body of text with numerical data series used to describe financial performance as it evolves over time. Our deep learning-based system pulls out from large amounts of textual data potentially relevant and useful textual descriptions that explain the performance of a financial metric of interest – without the need of human experts or labelled data.