O'Reilly on Data

DECEMBER 12, 2019

More structured approaches to sensitivity analysis include: Adversarial example searches : this entails systematically searching for rows of data that evoke strange or striking responses from an ML model. Figure 1 illustrates an example adversarial search for an example credit default ML model.

Machine Learning 276

Machine Learning Modeling Testing Risk Management 276

O'Reilly on Data

MARCH 31, 2020

Pragmatically, machine learning is the part of AI that “works”: algorithms and techniques that you can implement now in real products. We won’t go into the mathematics or engineering of modern machine learning here. After training, the system can make predictions (or deliver other results) based on data it hasn’t seen before.

Management 363

Management Machine Learning Experimentation Metrics 363

IBM Big Data Hub

NOVEMBER 29, 2023

For instance, if a business prioritizes accuracy in generating synthetic data, the resulting output may inadvertently include too many personally identifiable attributes, thereby increasing the company’s privacy risk exposure unknowingly. How to get started with synthetic data in watsonx.ai

Metrics 80

Metrics Machine Learning Statistics Risk 80

Sisense

APRIL 15, 2021

From data preparation , with attendant data quality assessment, to connecting to datasets and performing the analysis itself, helpful AI elements, invisibly integrated into the platform, make analysis smoother and more intuitive.

Analytics 98

Analytics Data Science Machine Learning Modeling 98

Smarten

APRIL 19, 2024

The value of an AI-focused analytics solution can only be fully realized when a business has ensured data quality and integration of data sources, so it will be important for businesses to choose an analytics solution and service provider that can help them achieve these goals.

Analytics 59

Analytics Predictive Modeling KPI Machine Learning 59

DataRobot Blog

MAY 26, 2022

These methods provided the benefit of being supported by rich literature on the relevant statistical tests to confirm the model’s validity—if a validator wanted to confirm that the input predictors of a regression model were indeed relevant to the response, they need only to construct a hypothesis test to validate the input.

Risk 52

Risk Modeling Metrics Business Objectives 52

Domino Data Lab

JUNE 23, 2019

It used deep learning to build an automated question answering system and a knowledge base based on that information. It is like the Google knowledge graph with all those smart, intelligent cards and the ability to create your own cards out of your own data.

Management 110

Management Machine Learning Experimentation Metrics 110

Domino Data Lab

MARCH 3, 2019

O’Reilly Media had an earlier survey about deep learning tools which showed the top three frameworks to be TensorFlow (61% of all respondents), Keras (25%), and PyTorch (20%)—and note that Keras in this case is likely used as an abstraction layer atop TensorFlow. The data types used in deep learning are interesting.

Data Science 49

Data Science Deep Learning Machine Learning Modeling 49

Domino Data Lab

JULY 22, 2019

He was saying this doesn’t belong just in statistics. He also really informed a lot of the early thinking about data visualization. It involved a lot of interesting work on something new that was data management. To some extent, academia still struggles a lot with how to stick data science into some sort of discipline.

Data Science 109

Data Science Machine Learning Data Governance Modeling 109

O'Reilly on Data

MARCH 24, 2020

” One of his more egregious errors was to continually test already collected data for new hypotheses until one stuck, after his initial hypothesis failed [4]. You may picture data scientists building machine learning models all day, but the common trope that they spend 80% of their time on data preparation is closer to the truth.

Machine Learning 364

Machine Learning Statistics Data Quality Modeling 364