Agile Machine Learning: Effective Machine Learning Inspired by the Agile Manifesto

ISBN-13: 9781484251065 / Angielski / Miękka / 2019 / 248 str.

Eric Carter; Matthew Hurst

Opis

Intermediate user level

Spis treści

Chapter 1: Early Delivery

Chapter 2: Changing Requirements

Chapter 3: Continuous Delivery

Chapter 4: Aligning with the Business

Chapter 5: Motivated Individuals

Chapter 6: Effective Communication

Chapter 7: Monitoring

Chapter 8: Sustainable Development 

Chapter 9: Technical Excellence

Chapter 10: Simplicity 

Chapter 11: Self-organizing Teams

Chapter 12: Tuning and Adjusting 

Chapter 13: Conclusion 

 

O autorze

Eric Carter has worked as a Partner Group Engineering Manager on the Bing and Cortana teams at Microsoft. In these roles he worked on search features around products and reviews, business listings, email, and calendar. He currently works on the Microsoft Whiteboard product.

Matthew Hurst is a Principal Engineering Manager and Applied Scientist currently working in the Machine Teaching group at Microsoft. He has worked in a number of teams in Microsoft including Bing Document Understanding, Local Search and in various innovation teams.

Blurb

Build resilient applied machine learning teams that deliver better data products through adapting the guiding principles of the Agile Manifesto.

Bringing together talented people to create a great applied machine learning team is no small feat. With developers and data scientists both contributing expertise in their respective fields, communication alone can be a challenge. Agile Machine Learning teaches you how to deliver superior data products through agile processes and to learn, by example, how to organize and manage a fast-paced team challenged with solving novel data problems at scale, in a production environment.

The authors’ approach models the ground-breaking engineering principles described in the Agile Manifesto. The book provides further context, and contrasts the original principles with the requirements of systems that deliver a data product.

What You'll Learn:

• Effectively run a data engineering team that is metrics-focused, experiment-focused, and data-focused
  
• Make sound implementation and model exploration decisions based on the data and the metrics
  
• Know the importance of data wallowing: analyzing data in real time in a group setting
  
• Recognize the value of always being able to measure your current state objectively
  
• Understand data literacy, a key attribute of a reliable data engineer, from definitions to expectations
  

This book is for anyone who manages a machine learning team, or is responsible for creating production-ready inference components. Anyone responsible for data project workflow of sampling data; labeling, training, testing, improving, and maintaining models; and system and data metrics will also find this book useful. Readers should be familiar with software engineering and understand the basics of machine learning and working with data.