In the News
Google Dethrones NVIDIA In Latest AI Benchmarking Tests
MLCommons - an open engineering consortium - conducted tests to determehow long it takes to train various neural networks. It concluded that Google’s cloud based TPU v4 ML supercomputer outperformed NVIDIA A100 in four out of eight training tests covering language (2), computer vision (4), reinforcement learning (1), and recommender systems (1).
Designing customer conversations: Best practices
A well designed virtual agent can increase customer self-service rates, improve customer satisfaction and reduce interaction handle times. With this handbook, learn how to put tried and tested conversation design principles into practice.
In The News
Tesla loses top AI executive who led Autopilot vision team
Andrej Karpathy, the deep learning and computer vision expert who was hired five years ago as Tesla’s director of AI and led Autopilot vision team, is officially leaving the company.
Humans Are The Future Of AI
Kevin Collins is CEO of Charli AI, a leading Intelligent Content Services Platform provider that is building AI that makes work flow.
A year in the making, BigScience’s AI language model is finally available
After more than a year of planning and training, a volunteer-led project has produced an open source language model that they claim is as powerful as OpenAI’s GPT-3, but free and open for anyone to use
Applied use cases
Why business is booming for military AI startups
The invasion of Ukraine has prompted militaries to update their arsenals—and Silicon Valley stands to capitalize.
Meta's New Bot ‘Sphere’ Can Crawl Through Wikipedia to Certify Citations
Meta's new AI is reportedly grabbing its information from 134 million public pages to aid Wikipedia's 6.5 million articles.
AI model finds potential drug molecules a thousand times faster
MIT researchers developed a geometric deep-learning model called EquiBind that is 1,200 times faster than one of the fastest existing computational molecular docking models, QuickVina2-W, in successfully binding drug-like molecules to proteins.
How can we protect members of algorithmic groups from AI-generated unfair outcomes?
AI creates unintuitive and unconventional groups to make life-changing decisions, yet current laws do not protect people from AI-generated unfair outcomes. In conversation with Professor Sandra Wachter as she explains how to resolve this issue.
AI researchers building surveillance tech and deepfakes resist ethical concerns
The computer vision research community is behind on AI ethics, but it’s not just a research problem. Practitioners say the ethics disconnect persists as young computer vision scientists make their way into the ranks of corporate AI.
Inside a radical new project to democratize AI
A group of over 1,000 AI researchers has created a multilingual large language model bigger than GPT-3—and they’re giving it out for free.
Responsive soft robots inspired by sputtering ketchup bottle
Researchers in the field of soft robotics are therefore developing robots made from soft, flexible materials that can respond to changes in their environment without external guidance.
A self-aware robot taught itself how to use its body
For a demonstration, a group of researchers has created a robot that could learn, through practice, what its own form can do.
What B.C. researchers think robots can learn from therapists
According to British Columbia-based researchers, robots have a lot to learn from therapists.
Intuitive physics learning in a deep-learning model inspired by developmental psychology
In this paper, researchers address this gap between humans and machines by drawing on the field of developmental psychology.
Large-scale multi-label text classification
In this example, we will build a multi-label text classifier to predict the subject areas of arXiv papers from their abstract bodies.
Does AutoML work for diverse tasks?
This blog post is dedicated to two recent but related efforts that measure the field’s current effectiveness at achieving this goal: NAS-Bench-360 and the AutoML Decathlon.