Every year, just after CES in Las Vegas, Nev., EE Times travels across the Atlantic Ocean to the annual HiPEAC conference in Europe, which we think is one of the key European forums for experts in computer architecture, programming models, compilers, and operating systems for general-purpose, embedded, and cyber-physical systems.
While the focus might be very much on presenting the latest academic research (this year, there are three keynotes, 33 workshops, and three tutorial sessions), there is enough industry involvement to make it a good place for researchers to network, university spinouts to showcase their technologies, and companies like Arm, AMD, ASML, Google, and Huawei, among others, to recruit talent.
As Koen De Bosschere, the coordinator of the HiPEAC project European, as well as the head of the Department of Electronics and Information Systems at Ghent University, puts it, “The HiPEAC conference acts as a bridge across the computing stack, amplifying the impact of research and nurturing future computing leaders.”
The conference rotates locations around Europe, and this year, it is being hosted in Krakow, Poland. On day one, the keynote talk held by Michaela Blott of AMD focused on enabling efficient AI through silicon diversity, model optimization, agile AI stacks, and other innovative approaches. In the video interview below, you can hear what she had to say, as well as catch our thoughts on Blott’s opening keynote, plus some of the other sessions we attended during the day.
www.eic.net.cn
The conference rotates locations around Europe, and this year, it is being hosted in Krakow, Poland. On day one, the keynote talk held by Michaela Blott of AMD focused on enabling efficient AI through silicon diversity, model optimization, agile AI stacks, and other innovative approaches. In the video interview below, you can hear what she had to say, as well as catch our thoughts on Blott’s opening keynote, plus some of the other sessions we attended during the day.
One of the key messages we took away from Blott’s opening talk was that there was a huge element of laziness in the community and industry as far as AI efficiency is concerned. It’s vital, she said, to pay attention to optimization at all levels to tackle some of the big challenges in AI becoming more efficient. “New algorithms are needed to bring AI efficiency in line with human performance and provide sustainable scaling,” Blott said. “Current methods are just too lazy.”
Blott’s message to the community was: “Don’t stop optimizing. Explore and co-design architectures with new algorithms, and in tandem with this, design better AI algorithms with better scaling properties.”
www.eic.net.cn
Technology meets society: cutting through the humanoid robotics hype
The great thing about listening to researchers who’ve actually done some real-world experiments with the technologies we talk about is that they inject a dose of reality. That was the case as we attended the “Technology Meets Society” track at HiPEAC 2026.
EE Times spoke to Micol Spitale, assistant professor at the Department of Electronics, Information, and Bioengineering at the Politecnico di Milano, Italy, who gave a talk on “socially intelligent robots” as part of this track, in which she explained that all the discussion about humanoid robotics is hyped up way too much. The reality is that there’s still a long way to go to meet the expectations that have been talked about for humanoid robotics.
In her work, she carried out research on groups of both children and adults in separate experiments to explore how support robots can be used to help promote human well-being.
In the following video interview, you’ll be able to understand more about her work and the potential for human-robot interaction to bring about technology for good:
PolyMage Labs, automatic compilers, and Tenstorrent partnership
Earlier this month, Indian startup PolyMage Labs announced that its automatic compiler for AI hardware was chosen by Tenstorrent for its AI platforms to improve software support for advanced AI hardware. PolyMage Labs said the partnership addresses a major challenge in the AI hardware industry: the lack of a mature and seamless software ecosystem, which significantly hinders the adoption of several AI accelerator platforms.
Uday Bondhugula, founder and CTO of Polymage Labs and a professor of computer science at the Indian Institute of Science in Bangalore, told EE Times more about the Tenstorrent partnership, as well as how the company is addressing current industry challenges to enable automated just-in-time compilation to support multiple AI frameworks. You can watch the video interview below:
www.eic.net.cn
See also:
易IC库存管理软件
RELATED TOPICS:
Share this:
Click to share on X (Opens in new window)
X
Facebook
LinkedIn
Follow Nitin on LinkedIn
www.eic.net.cn
易IC库存管理软件
