French startup LightOn has raised $3.3 million in seed funding for a new kind of optics-based hardware technology that improves data processing for artificial intelligence.
Leveraging compressive sensing, LightOn said its hardware and software can make AI computations both simpler and orders of magnitude more efficient. The technology, licensed by PSL Research University, was originally developed at several of Paris’ leading research institutions.
For the past few months, LightOn has allowed a select group of beta customers access to its Optical Processing Units (OPU) through the LightOn Cloud, thanks to a partnership with OVH, Europe’s leading cloud provider. First users from both academia and industry have already successfully demonstrated impressive results on this hybrid CPU/GPU/OPU server, which has outperformed silicon-only computing technology in a variety of large-scale machine learning tasks.
The Paris-based company has raised money from technology-focused venture funds, including Quantonation and Anorak, which is headed by Greg Castle. Christophe Jurczak, managing partner at Quantonation, will join LightOn’s board.
Typical use cases currently include transfer learning, change point detection, and time series prediction.
LightOn’s CEO, Igor Carron, runs a large European meetup for AI and popularizes his work through the AI research blog Nuit Blanche.
“It’s an exciting time as artificial intelligence develops rapidly,” Carron said in a statement. “The requirements as usage scales necessitate improved power efficiency and performance. LightOn’s technology addresses these monumental challenges.”
Carron is joined by three cofounders with academic recognition: Laurent Daudet, now CTO at LightOn and an expert in signal processing; Sylvain Gigan, who provides the optics expertise; and Florent Krzakala, who coordinates the machine learning efforts at LightOn.
“We’re very excited at Quantonation to have the opportunity through this investment to support an amazing founding team with extremely impressive academic credentials, as well as a deep understanding of the challenges facing current computing paradigms for Machine Learning,” said Jurczak, in a statement. “We believe that any solution that is not energy-efficient will face huge difficulties to scale at some point, and that’s one of LightOn’s key value propositions. Their technology is already available and will for sure be a key enabler present in future HPC devices on premise, as well as cloud-enabled servers for machine learning.”
This seed funding will accelerate both core technology development and the LightOn Cloud product to bring LightOn’s optical technology in large scale machine learning computations to the AI community. Developers interested in testing the capabilities of the new platform for potential deployment can visit LightOn.io and join the wait list.
The company was previously funded by the founders and government grants, and it has 10 employees.