[PDF][PDF] Introduction
RRM Coleman - Say It Loud!, 2013 - library.oapen.org
Now we demand a chance to do things for ourself We're tired ofbeatin our head against the
wall Say it loud, I'm Black and I'm proud.—James Brown," Say It Loud" The past three …
wall Say it loud, I'm Black and I'm proud.—James Brown," Say It Loud" The past three …
Motivation and Best Practices for Machine Learning Designers and Testers
This chapter highlights best practices and pitfalls in the development of ML systems.
Following the best practices and avoiding the pitfalls presented should increase the …
Following the best practices and avoiding the pitfalls presented should increase the …
Sequential Drift Detection
This chapter discusses drift detection in the context of data observed in a time-ordered
sequence, as opposed to the static datasets in Chaps. 7 and 8. It covers concepts including …
sequence, as opposed to the static datasets in Chaps. 7 and 8. It covers concepts including …
Drift Detection by Measuring Distribution Differences
This chapter discusses the concepts of populations and samples of observational units and
the idea of drift detection as being an inference that drift has occurred in a population …
the idea of drift detection as being an inference that drift has occurred in a population …
A Framework Analysis for Alternating Components and Drift
This chapter discusses a framework architecture that helps choose the best design of an ML
system based on the data and models that compose the system. As data and business goals …
system based on the data and models that compose the system. As data and business goals …
Principles of Drift Detection and ML Solution Retraining
This chapter introduces the concept of drift as a change in the statistical distributions of
observed data or in the performance of an ML model. Several representative types of drifts …
observed data or in the performance of an ML model. Several representative types of drifts …
Testing Solutions Based on Large Language Models
Large language models (LLMs) are ML models trained on extensive data corpora, enabling
them to perform multiple downstream ML tasks. Consequently, using LLMs allows the ML …
them to perform multiple downstream ML tasks. Consequently, using LLMs allows the ML …
Drift in Characterizations of Data
This chapter presents several examples from the authors' research on applications of drift
detection in industrial settings. These include representing a dataset as “slices” based on …
detection in industrial settings. These include representing a dataset as “slices” based on …
Scientific Analysis of ML Systems
The primary requirement of a good learning system is generalization, ie, the ability to
generate rules that perform well on data similar to the training data. This is achieved through …
generate rules that perform well on data similar to the training data. This is achieved through …
Optimal Integration of the ML Solution in the Business Decision Process
The ML system is integrated into a business process to help increase the value obtained
from the process by the organization. Once the probability of error by the ML system is …
from the process by the organization. Once the probability of error by the ML system is …