Prestigious fellowship opens doors for Wits AI expert

When Benji Rosman first started tinkering with robots as a young computer science student at the University of the Witwatersrand (Wits), he never imagined that his curiosity would one day place him among the world’s most influential thinkers in artificial intelligence (AI).

That journey reached a new milestone on 21 July, when Rosman was named a fellow of the Canadian Institute for Advanced Research (CIFAR) Learning in Machines & Brains programme, an honour reserved for only a handful of leading researchers worldwide.

Being appointed a fellow of the CIFAR programme is a prestigious recognition of a researcher’s exceptional contributions to understanding intelligence through interdisciplinary approaches. The programme unites experts from neuroscience, AI, and related fields to explore how biological and artificial systems acquire and process knowledge.

CIFAR connects researchers from different countries, disciplines, and institutions into specialised programmes. These programmes focus on fields such as AI, quantum materials, child development, brain science, and climate solutions.

In 2022, Rosman was one of 18 early-career researchers in the world to be selected for the prestigious CIFAR Azrieli Global Scholars programme for 2022-2024, which afforded him unrestricted research support to pursue high-risk, high-reward ideas.

Now, following the end of his tenure with the programme, Rosman was shocked to hear that he had been named a fellow of the CIFAR Learning in Machines & Brains programme, as he didn’t apply for it.

“I’ve been working with them over the past couple of years, but to be given this next level, somewhat more permanent role, was a big deal to me,” he said. “This is one of the most prestigious groups in machine learning and AI that has been driving the development and evolution of the field for decades.”

The Learning in Machines & Brains programme is one of CIFAR’s flagship initiatives. It gathers leading AI researchers, like Rosman to advance understanding of machine learning and neuroscience and to develop cutting-edge AI methods.

“It’s a group of some of the core people around the world working in advances in machine learning or what’s called AI, and also computational neuroscience,” he said.

Rosman said working with CIFAR was coveted in his field, so when he was given a foot in the door, he knew he couldn’t turn it down.

“CIFAR has done leading work, not just in AI, but in everything from physics to biosciences, to social sciences. The opportunity to be in a room with – and work with – authority leaders, particularly in the Global Scholars Programme, is the kind of opportunity one dreams of in science and academia – for us to be able to collaborate with this kind of person and work on real projects that could have a global impact,” he said.

The scientists will hold regular meetings and workshops where they will collaborate with peers from diverse disciplines to foster innovative insights into learning mechanisms in brains and machines.

Rosman, the founding director of the Wits Machine Learning and Neural Discovery (MIND) Institute, a flagship institute for artificial and natural intelligence in Africa, said the fellowship would give the institute more opportunities to share what it is doing on a global stage.

“The MIND Institute is always looking for different ways of partnering to make sure that our work has maximal global impact, and this provides opportunities to make sure the work we’re doing gets more visibility globally,” he said. “It’s great to be involved in working with some of the other thought leaders and pushing our agenda as well, which is around building the next generation of AI technology from Africa itself.”

As part of his fellowship, Rosman has access to support for financial research. However, he said being a part of CIFAR is about more than financing research; it’s about the connections that can be made with collaborators around the globe.

“They give us some kind of small amount of financial-research support. But, you know, also it’s a network with a lot of connections and opportunities for pushing other initiatives, whether they’re around fundraising or policy work. Having the profile they have, they’re just so well-connected and super useful as well,” he said.

“There are projects and programmes that I’m involved in now that have come about directly as a result of collaborations in CIFAR,” he said, “ranging from some of the work that we’re doing in looking at things around machine consciousness; looking at the impact of AI; or the interface of AI and possible connections to climate challenges. These kinds of things have become a bigger part of my work because of the CIFAR network.”