Microsoft goes all in on Azure Quantum to accelerate scientific discovery

In the rapidly evolving landscape of quantum computing, Microsoft is pushing the boundaries with its Azure Quantum platform. As Microsoft CEO Satya Nadella recently announced, “Our goal is to compress the next 250 years of chemistry and material science progress into the next 25,” Nadella said. The executive also noted that, thanks to recent advances in quantum computing, “we’re closer than ever to the promise of a scalable quantum machine capable of solving some of the world’s most intractable problems.”

To accelerate this vision, Microsoft’s Azure Quantum introduces capabilities to expedite scientific discovery. Azure Quantum Elements merges high-performance computing, AI and quantum computing on Microsoft Azure. This combination accelerates quantum chemistry simulations and widens the search space for new materials. Azure Quantum Elements taps AI models trained on millions of data points to significantly accelerate quantum chemistry simulations. In an internal Microsoft study comparing ab initio molecular dynamics and active learning Molecular Dynamics screening methods, Azure Quantum Elements achieved up to 500,000 fold faster simulations.

Azure Quantum Elements, a high-performance computing system that merges AI and quantum computing, aims to slash R&D pipeline timelines and costs. “Even a simple chemical reaction can involve millions of possible steps, which must be analyzed often in the form of a network. This is an extremely challenging computational process that demands the scale of Azure,” said Nihit Pokhrel, a computational chemist on the Azure Quantum team in a virtual briefing.

Copilot in Azure Quantum

Similar to how GitHub Copilot aids software developers in writing code, Copilot in Azure Quantum assists researchers in complex scientific workflows. Scientists can navigate quantum chemistry and materials science problems through natural language conversations with Copilot.

Going beyond conversational interactions, Copilot can automatically generate calculations, run simulations, query data and offer guided explanations to users. These capabilities reduce friction in research workflows by enabling scientists to accomplish tasks more seamlessly without switching between tools.

Microsoft’s quantum computing roadmap

To track progress in quantum computing, Microsoft has established three Quantum Computing Implementation Levels:

1. Foundational. The first stage mirrors early classical computing with noisy qubits like vacuum tubes that are hard to scale. This level includes today’s Noisy Intermediate Scale Quantum (NISQ) computers. As Zander explains, “The foundational stage, level one, mirrors the early days of classical computing. Current qubits, much like these initial systems, are noisy and challenging to scale.”
2. Resilient. This stage moves from noisy qubits to reliable logical qubits, critical for scalable quantum computers. “Transitioning into level two, we see the emergence of resilience. This stage is marked by the move from noisy physical qubits to reliable logical qubits, a critical step towards creating a scalable quantum supercomputer,” Zander notes. Logical qubits built on Majorana particles represent a significant recent advance toward this stage.
3. Scale. The final level focuses on performance measured in reliable quantum operations per second.  Zander continues, “Finally, level three, scale, necessitates a shift in our approach. Rather than focusing on the number of qubits, we introduce a new performance metric: reliable Quantum Operations Per Second (rQOPS).”

Competitive landscape in quantum and quantum-inspired computing

Microsoft is not the only player in the field of quantum computing. In June, Alphabet spinoff SandboxAQ recently unveiled the biopharma molecular simulation division AQBioSim. Additionally, Accenture and quantum software firm 1QBit have partnered with Biogen on a quantum-enabled molecular comparison app to potentially speed up drug discovery for complex neurological conditions. AWS created an open-source Quantum Computing Exploration for Drug Discovery solution on its cloud platform. IBM is also developing a hybrid cloud and quantum computing environment to expedite drug discovery. Other organizations tapping quantum-inspired technologies include Fujitsu, Boehringer Ingelheim and Novo Nordisk.

As Microsoft corporate vice president Jason Zander explains, “The journey to quantum computing parallels the evolution of classical computing, starting from a foundational level and advancing through resilience and scale. As we navigate these stages, we remain focused on our ultimate goal: a general-purpose, programmable quantum supercomputer.”

---

Filed Under: Data science, Drug Discovery, Drug Discovery and Development, Industry 4.0, machine learning and AI
Tagged With: AI in Pharma, drug discovery, high-performance computing, Microsoft Azure, molecular simulations, quantum chemistry, quantum computing
 

---