Get to the heart of real quantum hardware

Our newest freely available quantum computing system takes one more step toward bringing the lab to the cloud. It features pulse-level control, and when coupled with today’s release of the new version of Qiskit (version 0.14), any IBM Quantum Experience user now has the ability to construct schedules of pulses and execute them. The role of experimental quantum physicist is now available to anyone with internet access.

Continue reading

Extending 8-bit training breakthroughs to the toughest deep learning models

At NeurIPS 2019, IBM Research continues to advance its 8-bit training platform to improve performance and maintain accuracy for the most challenging emerging deep learning models.

Continue reading

Addressing open fundamental questions in reinforcement learning

New IBM research presented at NeurIPS 2019 introduces an innovative probabilistic framework for reinforcement learning that helps address three open questions central to the technology.

Continue reading

IBM Q Network Adds New Member, Stanford University’s Q-Farm Initiative

At the Q2B 2019 Conference, IBM announced that Stanford University’s Q-Farm initiative, a collaborative with the SLAC National Accelerator Laboratory, has joined the IBM Q Network. As a member organization, Q-FARM will collaborate with IBM to accelerate joint research in quantum computing and develop curricula to help prepare students for careers that will be influenced by this next era of computing across science and business.

Continue reading

The path to the “perfect” analog material and system: IBM at IEDM and NeurIPS

Researchers from the IBM AI Hardware Center will showcase at IEDM and NeurIPS new analog devices, algorithmic and architectural solutions, a novel model training technique, and a full custom design.

Continue reading

Nanosheet Technology for the Computing Era of AI and 5G

At IEDM, the top conference for semiconductor device technology, IBM Research presents the latest progress in nanosheet technology, including new critical features for high performance computing.

Continue reading

Quantum-Inspired Logical Embedding for Knowledge Representation

In our new paper, to be presented at NeurIPS 2019, we develop a new knowledge representation, which we call “quantum embedding”, that represents conceptual knowledge using a vector space representation that preserves its logical structure and allows reasoning tasks to be solved accurately and efficiently.

Continue reading

Learning and Characterizing Causal Graphs with Latent Variables

Researchers from the MIT-IBM Watson AI Lab developed a new approach to characterize the set of plausible causal graphs from observational and interventional data that has latent variables.

Continue reading

Enveritas Pilots IBM’s AI-powered AgroPad to Help Coffee Farmers

AI-powered technology will aid smallholder coffee farmers and help reduce global poverty in the coffee sector by 2030 A smart, AI-powered paper device the size of a business card dubbed the IBM AgroPad can remotely and quickly analyze soil samples for chemical composition – and could help smallholder coffee farmers save money and improve sustainability. […]

Continue reading

Harnessing Gallium Phosphide for Future Information Technology

In the paper “Integrated gallium phosphide nonlinear photonics”, recently published in the peer-reviewed journal Nature Photonics, we report on the development of high-performance photonic devices made of the crystalline semiconductor gallium phosphide. This work represents a breakthrough in the manipulation of light with semiconductor materials integrated on a chip. It opens the door to a […]

Continue reading

IBM Research Develops AI to Greatly Improve Tea Inspection

When you drink that cup of tea, you sure don’t want any fluff floating there. That’s the most critical step in tea production – removing impurities, be it stems, seeds, stones or even bugs. If they do make it into your drink, the brand may get fined; at the very least, you probably won’t buy […]

Continue reading

More Is Less: Learning Efficient Video Representations

IBM researchers developed a novel low memory footprint and efficient architecture for spatio-temporal analysis of video. The results show strong performance on several benchmarks – and allow training of deeper models using larger sequences of input frames, which will lead to higher accuracy on video action recognition tasks.

Continue reading