In June, Microsoft available new Azure Machine Learning (ML) courses at Udacity, providing interested parties with open source tools and frameworks such as PyTorch for building complex ML solutions. Last week at Ignite, Microsoft Power BI CVP Arun Ulag mentioned the importance of Azure ML services in construction end-to-end systems.
Today, biopharmaceutical company AstraZeneca released information about how uses Azure ML and PyTorch to accelerate research into new drug development.
The team believes that machine learning is the key to analyzing data to find relevant connections, so it used a knowledge-based graph approach to understand the relationship between contextual scientific data networks.
Its processing members physicsof language (NLP from natural language processing) preferred the PyTorch framework for building various models, following the latest research.
In conjunction with Azure ML, they created proposition systems that can be used to map nodes in the knowledge graph to numerical data. These systems are then used to train the use of specific case models in a coherent manner, through the advanced computing capabilities of the Azure ML.
A recent document AstraZeneca compares the performance of the model under different conditions. The company uses Azure Blob storage to handle the vast amounts of data required.
Similarly, the management end-to-end lifecycle for the entire machine learning process is also facilitated through Azure ML, as shown in the figure above, speeding up iterations and the model development process.
Models built using these approaches are ultimately used to discover and suggest "potential new and new drug targets" in a faster and more accurate way.
In the future, the company plans to continue to expand its knowledge chart - applying machine learning through the aforementioned platforms, with the ultimate goal of building new drugs in the healthcare industry more efficiently.