Cassie Kozyrkov, Chief Decision Scientist at Google Cloud, introduces the concept of Decision Intelligence and the combination of behavioral and data sciences. She discusses the importance of understanding decision context and utilizing data while exploring challenges in AI and machine learning.
Decision intelligence combines data science with behavioral and managerial sciences to ensure wise and informed decision making.
Splitting data into exploratory and test sets enables stronger data-driven decision making and helps avoid bias.
Deep dives
TPUs now available in beta on cloud machine learning engine and Kubernetes 1.10 now generally available
TPUs are now available in beta on cloud machine learning engine, allowing users to play around with TPUs without needing approval. Additionally, Kubernetes 1.10 is now generally available with new features such as shared virtual private cloud, node auto repairs, regional persistent disks, and more.
Ultraman Google Compute Engine Machine types and App Engine Scheduler's performance improvements
Ultraman Google Compute Engine Machine types, family of memory optimized virtual machine instances with large amounts of memory, are now available. They provide up to 3,844 gigabytes of memory and 160 CPUs, making them ideal for memory-intensive tasks. In addition, the new App Engine Scheduler improves cold starts by 30%, reduces costs by 7%, and shows an average 5% reduction in request latencies.
Cassie Kosrickoff shares insights on decision intelligence
In this episode, Cassie Kosrickoff, Chief Decision Scientist for Google Cloud, discusses decision intelligence. She explains that decision making is crucial for businesses and individuals as it affects how they interact with the world. Decision intelligence combines data science with behavioral and managerial sciences to ensure wise and informed decision making. Cassie emphasizes the importance of framing decision context upfront, measuring success, and using a methodical process to approach decision making.
The importance of splitting data for machine learning
Cassie Kosrickoff highlights the significance of splitting data when utilizing machine learning. By splitting data into exploratory and test sets, organizations can engage employees across all education levels in discovering insights and getting inspired from the exploratory set. Decision makers can then use this inspiration to frame their decision context and rigorously evaluate it using the test set. The split helps to avoid bias and enables stronger data-driven decision making.
Chief Decision Scientist, Cassie Kozyrkov joins Mark and Melanie this week to explain data science, analytics, machine learning and statistical inference, in relation to decision intelligence.
Cassie Kozyrkov
As Chief Decision Scientist at Google Cloud, Cassie advises leadership teams on decision process, AI strategy, and building data-driven organizations. She works to democratize statistical thinking and machine learning so that everyone - Google, its customers, the world! - can harness the beauty and power of data. She is the innovator behind the practice of Decision Intelligence Engineering at Google and she has personally trained over 15,000 Googlers in machine learning, statistics, and data-driven decision-making. Before her current role, she served in Google’s Office of the CTO as Chief Data Scientist. Prior to joining Google, Cassie worked as a data scientist and consultant. She holds degrees in mathematical statistics, economics, psychology, and cognitive neuroscience. When she’s not working, you’re most likely to find Cassie at the theater, in an art museum, exploring the world, or curled up with a good novel.
Cool things of the week
Cloud ML Engine adds Cloud TPU support for training blogdocs
Google Kubernetes Engine 1.10 is generally available and ready for the enterprise blog
Introducing ultramem Google Compute Engine machine types blog
Increase performance while reducing costs with the new App Engine scheduler blog docs