Ericsson Launches Cognitive Labs to Advance AI for Telecoms

Ericsson has launched Cognitive Labs, a research-driven initiative dedicated to advancing AI for telecoms. The initiative operates virtually, exploring AI technologies such as Graph Neural Networks (GNNs), Active Learning, and Large-Scale Language Models (LLMs).

A Triad of Ericsson AI Labs

Central to the Cognitive Labs initiative are three distinct research arms, each focused on a specialized area of AI:

• GAI Lab (Geometric Artificial Intelligence Lab): This lab explores Geometric AI, emphasizing explainability in geometric learning, graph generation, and temporal GNNs. Ericsson anticipates this work will make significant contributions to areas like drug discovery and enable the creation of more efficient pharmaceuticals.
• MLR Lab (Machine Learning and Reasoning Lab): Focusing on training model optimization and reinforcement learning, this lab aims to advance energy-efficient training for AI models and support the creation of digital twins that simulate physical realities. The potential benefits include reductions in time and energy costs, key priorities in sustainable AI development.
• FAI Lab (Fundamental Artificial Intelligence Lab): Dedicated to foundational AI models like Large-Scale Language Models, this lab seeks to shape the future of AI applications in telecoms. Its primary mission is to automate processes and improve technological efficiencies crucial to the sector.

Open-Source Collaboration

In line with its ties to the open-source community, Cognitive Labs prioritizes the development of reusable AI libraries. This accessibility helps both researchers and industry professionals accelerate progress across multiple domains—from healthcare to digital communications.

Ericsson hopes to attract leading global researchers and data scientists to its fold, enhancing its credentials as a leader in AI innovation.

Partnership with Universidad Pontificia Comillas

Spain is emerging as a cornerstone for Cognitive Labs, with Ericsson Spain playing a leading role due to its robust R&D activities in Madrid and Málaga. A new agreement with Universidad Pontificia Comillas, specifically with its School of Engineering (Comillas ICAI), has been bolstered by this focus.

The alliance between Ericsson and ICAI is set to accelerate AI research through joint publications and active participation in open-source initiatives. Teams of Ericsson’s top engineers and data scientists will closely collaborate with the university’s research teams to drive innovation at both national and European levels.

Conclusion

Cognitive Labs marks a vital step forward for Ericsson and industries and communities that stand to benefit from its far-reaching, transformative potential. With AI set to play a critical role in shaping the next generation of mobile communications, this initiative underscores Ericsson’s commitment to leading innovation in AI and making a significant contribution to the open-source community.

See Also

• World Economic Forum unveils blueprint for equitable AI
• AI & Big Data Expo
• Intelligent Automation Conference
• BlockX
• Digital Transformation Week
• Cyber Security & Cloud Expo