- Coding Ethics for AI & AIOps: Designing Responsible AI Systems - Aug 26, 2021.
AI ops has taken Human machine collaboration to the next level where humans and machines are not just coexisting but are collaborating and working together like team members.
- What is Noise? - Aug 25, 2021.
We might have a reasonable sense for what "noise" is as some statically random phenomena that occurs in Nature. But, how can this same characteristic be defined--and understood--within the context of making judgements, such as in human behavior, corporate decision-making, medicine, the law, and AI systems?
- Demystifying AI: The prejudices of Artificial Intelligence (and human beings) - Aug 20, 2021.
AI models are necessarily trained on historical data from the real-world--data that is generated from the daily goings on of society. If social-based biases are inherent in the training data, then will the AI predictions highlight these same biases? If so, what should we do (or not do) about making AI fair?
- Visualizing Bias-Variance - Aug 10, 2021.
In this article, we'll explore some different perspectives of what the bias-variance trade-off really means with the help of visualizations.
- Towards a Responsible and Ethical AI - Jul 30, 2021.
It is not the technology at fault, but the intention.
- KDnuggets™ News 21:n27, Jul 21: Top 6 Data Science Online Courses in 2021; Geometric Foundations of Deep Learning - Jul 21, 2021.
Top 6 Data Science Online Courses in 2021; Geometric foundations of Deep Learning; Google’s Director of Research Advice for Learning Data Science; SQL, Syllogisms, and Explanations; How to Create Unbiased Machine Learning Models
- How to Create Unbiased Machine Learning Models - Jul 16, 2021.
In this post we discuss the concepts of bias and fairness in the Machine Learning world, and show how ML biases often reflect existing biases in society. Additionally, We discuss various methods for testing and enforcing fairness in ML models.
- Ethics, Fairness, and Bias in AI - Jun 30, 2021.
As more AI-enhanced applications seep into our daily lives and expand their reach to larger swaths of populations around the world, we must clearly understand the vulnerabilities trained machine leaning models can exhibit based on the data used during development. Such issues can negatively impact select groups of people, so addressing the ethical decisions made by AI--possibly unknowingly--is important to the long-term fairness and success of this new technology.
- What Makes AI Trustworthy? - May 11, 2021.
This blog pertains to the importance of why AI needs to be trustworthy as well as what makes it trustworthy. AI predictions/suggestions should not just be taken at face value, but rather delved into at a deeper level. We need to understand how an AI system makes its predictions to put our trust in it. Trust should not be built on prediction accuracy alone.
- The Three Edge Case Culprits: Bias, Variance, and Unpredictability - Apr 22, 2021.
Edge cases occur for three basic reasons: Bias – the ML system is too ‘simple’; Variance – the ML system is too ‘inexperienced’; Unpredictability – the ML system operates in an environment full of surprises. How do we recognize these edge cases situations, and what can we do about them?
- Top 3 Statistical Paradoxes in Data Science - Apr 15, 2021.
Observation bias and sub-group differences generate statistical paradoxes.
- Popular Machine Learning Interview Questions - Jan 20, 2021.
Get ready for your next job interview requiring domain knowledge in machine learning with answers to these eleven common questions.
- How to easily check if your Machine Learning model is fair? - Dec 24, 2020.
Machine learning models deployed today -- as will many more in the future -- impact people and society directly. With that power and influence resting in the hands of Data Scientists and machine learning engineers, taking the time to evaluate and understand if model results are fair will become the linchpin for the future success of AI/ML solutions. These are critical considerations, and using a recently developed fairness module in the dalex Python package is a unified and accessible way to ensure your models remain fair.
- AI registers: finally, a tool to increase transparency in AI/ML - Dec 9, 2020.
Transparency, explainability, and trust are pressing topics in AI/ML today. While much has been written about why they are important and what you need to do, no tools have existed until now.
- 20 Core Data Science Concepts for Beginners - Dec 8, 2020.
With so much to learn and so many advancements to follow in the field of data science, there are a core set of foundational concepts that remain essential. Twenty of these ideas are highlighted here that are key to review when preparing for a job interview or just to refresh your appreciation of the basics.
- From Y=X to Building a Complete Artificial Neural Network - Nov 13, 2020.
In this tutorial, we will start with the most simple artificial neural network (ANN) and move to something much more complex. We begin by building a machine learning model with no parameters—which is Y=X.
- Six Ethical Quandaries of Predictive Policing - Nov 6, 2020.
When predictive machine learning models are applied to real-life scenarios, especially those that directly impact humans, such as cancer detection and other medical-related applications, the risks involved with incorrect predictions carry very high stakes. These risks are also prominent in how machine learning is applied in law enforcement, and serious ethical questions must be considered.
- Overcoming the Racial Bias in AI - Oct 30, 2020.
The results of any AI developed today is entirely dependent on the data on which it trains. If the data is distributed--intentionally or not--with a bias toward any category of data over another, then the AI will display that bias. What is a better way forward to handle this possibility toward bias when the datasets involve human beings?
- Can AI Learn Human Values? - Oct 27, 2020.
OpenAI believes that the path to safe AI requires social sciences.
- DeepMind Relies on this Old Statistical Method to Build Fair Machine Learning Models - Oct 23, 2020.
Causal Bayesian Networks are used to model the influence of fairness attributes in a dataset.
- The Ethics of AI - Oct 21, 2020.
Marketing scientist Kevin Gray asks Dr. Anna Farzindar of the University of Southern California about a very important subject - the ethics of AI.
- Understanding Bias-Variance Trade-Off in 3 Minutes - Sep 11, 2020.
This article is the write-up of a Machine Learning Lighting Talk, intuitively explaining an important data science concept in 3 minutes.
- Top Google AI, Machine Learning Tools for Everyone - Aug 18, 2020.
Google is much more than a search company. Learn about all the tools they are developing to help turn your ideas into reality through Google AI.
- Word Embedding Fairness Evaluation - Aug 5, 2020.
With word embeddings being such a crucial component of NLP, the reported social biases resulting from the training corpora could limit their application. The framework introduced here intends to measure the fairness in word embeddings to better understand these potential biases.
- Free From Stanford: Ethical and Social Issues in Natural Language Processing - Jul 17, 2020.
Perhaps it's time to take a look at this relatively new offering from Stanford, Ethical and Social Issues in Natural Language Processing (CS384), an advanced seminar course covering ethical and social issues in NLP.
- Bias in AI: A Primer - Jun 23, 2020.
Those interested in studying AI bias, but who lack a starting point, would do well to check out this introductory set of slides and the accompanying talk on the subject from Google researcher Margaret Mitchell.
- Five Cognitive Biases In Data Science (And how to avoid them) - Jun 12, 2020.
Everyone is prey to cognitive biases that skew thinking, but data scientists must prevent them from spoiling their work. Learn more about five biases that can all too easily make your seemingly objective work become surprisingly subjective.
- Machine Fairness: How to assess AI system’s fairness and mitigate any observed unfairness issues - May 26, 2020.
Microsoft is bringing the latest research in responsible AI to Azure (both Azure Machine Learning and their open source toolkits), to empower data scientists and developers to understand machine learning models, protect people and their data, and control the end-to-end machine learning process.
- Were 21% of New York City residents really infected with the novel coronavirus? - May 6, 2020.
Understanding the types of statistical bias that pop up in popular media and reporting is especially important during this pandemic where the data -- and our global response to the data -- directly impact peoples' lives.
- 5 Ways to Apply Ethics to AI - Dec 19, 2019.
Here are six more lessons based on real life examples that I think we should all remember as people working in machine learning, whether you’re a researcher, engineer, or a decision-maker.
- Dusting Under the Bed: Machine Learners’ Responsibility for the Future of Our Society - Dec 13, 2019.
The Machine Learning community must shape the world so that AI is built and implemented with a focus on the entire outcome for our society, and not just optimized for accuracy and/or profit.
- What just happened in the world of AI? - Dec 12, 2019.
The speed at which AI made advancements and news during 2019 makes it imperative now to step back and place these events into order and perspective. It's important to separate the interest that any one advancement initially attracts, from its actual gravity and its consequential influence on the field. This review unfolds the parallel threads of these AI stories over this year and isolates their significance.
- 5 Statistical Traps Data Scientists Should Avoid - Oct 30, 2019.
Here are five statistical fallacies — data traps — which data scientists should be aware of and definitely avoid.
- 3 Ways to Manage Human Bias in the Analytics Process - Sep 5, 2019.
Managing human bias is an important part of the analytics process. Learn about three areas to watch out for to ensure your models are as unbiased as possible.
- Python Libraries for Interpretable Machine Learning - Sep 4, 2019.
In the following post, I am going to give a brief guide to four of the most established packages for interpreting and explaining machine learning models.
- Types of Bias in Machine Learning - Aug 29, 2019.
The sample data used for training has to be as close a representation of the real scenario as possible. There are many factors that can bias a sample from the beginning and those reasons differ from each domain (i.e. business, security, medical, education etc.)
- Is Bias in Machine Learning all Bad? - Jul 23, 2019.
We have been taught over our years of predictive model building that bias will harm our model. Bias control needs to be in the hands of someone who can differentiate between the right kind and wrong kind of bias.
- “Please, explain.” Interpretability of machine learning models - May 9, 2019.
Unveiling secrets of black box models is no longer a novelty but a new business requirement and we explain why using several different use cases.
- 3 Big Problems with Big Data and How to Solve Them - Apr 18, 2019.
We discuss some of the negatives of using big data, including false equivalences and bias, vulnerability to security breaches, protecting against unauthorized access and the lack of international standards for data privacy regulations.
- Building NLP Classifiers Cheaply With Transfer Learning and Weak Supervision - Mar 15, 2019.
In this blog, I’ll walk you through a personal project in which I cheaply built a classifier to detect anti-semitic tweets, with no public dataset available, by combining weak supervision and transfer learning.
Pages: 1 2
- Designing Ethical Algorithms - Mar 8, 2019.
Ethical algorithm design is becoming a hot topic as machine learning becomes more widespread. But how do you make an algorithm ethical? Here are 5 suggestions to consider.
- Reflections on the State of AI: 2018 - Feb 26, 2019.
We provide a detailed overview of the key developments in the AI space, focusing on key players, applications, opportunities, and challenges.
- The Algorithms Aren’t Biased, We Are - Jan 29, 2019.
We explain the concept of bias and how it can appear in your projects, share some illustrative examples, and translate the latest academic research on “algorithmic bias.”
- Data Science and Ethics – Why Companies Need a new CEO (Chief Ethics Officer) - Jan 21, 2019.
We explain why data science companies need to have a Chief Ethics Officer and discuss their importance in tackling algorithm bias.
- 10 Exciting Ideas of 2018 in NLP - Jan 16, 2019.
We outline a selection of exciting developments in NLP from the last year, and include useful recent papers and images to help further assist with your learning.
- How to Remove Unfair Bias From Your AI - Jan 11, 2019.
In this live webinar, Jan 17 @ 11:00 am ET, Colin Priest, Senior Director of Product Marketing at DataRobot will discuss how to identify and correct bias in AI.
- A Case For Explainable AI & Machine Learning - Dec 27, 2018.
In support of the explainable AI cause, we present a variety of use cases covering operational needs, regulatory compliance and public trust and social acceptance.
- What If the Data Tells You to Be Racist? When Algorithms Explicitly Penalize - Sep 26, 2018.
Without the right precautions, machine learning — the technology that drives risk-assessment in law enforcement, as well as hiring and loan decisions — explicitly penalizes underprivileged groups.
- You Aren’t So Smart: Cognitive Biases are Making Sure of It - Sep 17, 2018.
Cognitive biases are tendencies to think in certain ways that can lead to systematic deviations from a standard of rationality or good judgment. They have all sorts of practical impacts on our lives, whether we want to admit it or not.
- The 2018 Data Scientist Report is Here - Aug 23, 2018.
Learn about the data and tools that data scientists are working with in 2018, Ethical issues around AI, Algorithmic bias, Job satisfaction, and more.
- Weak and Strong Bias in Machine Learning - Jul 6, 2018.
With the arrival of the GDPR there has been increased focus on non-discrimination in machine learning. This post explores different forms of model bias and suggests some practical steps to improve fairness in machine learning.
- Machine Learning Breaking Bad – addressing Bias and Fairness in ML models - May 25, 2018.
As the use of analytics proliferate, companies will need to be able to identify models that are breaking bad.
- Justice Can’t Be Colorblind: How to Fight Bias with Predictive Policing - Feb 28, 2018.
Predictive policing uncovers racial inequity, which it threatens to perpetuate - but, if we turn things around, it also presents an unprecedented opportunity to advance social justice.
- Error Analysis to your Rescue – Lessons from Andrew Ng, part 3 - Jan 29, 2018.
The last entry in a series of posts about Andrew Ng's lessons on strategies to follow when fixing errors in your algorithm
- Propensity Score Matching in R - Jan 18, 2018.
Propensity scores are an alternative method to estimate the effect of receiving treatment when random assignment of treatments to subjects is not feasible.
Pages: 1 2
- Learning Curves for Machine Learning - Jan 17, 2018.
But how do we diagnose bias and variance in the first place? And what actions should we take once we've detected something? In this post, we'll learn how to answer both these questions using learning curves.
Pages: 1 2
- Why understanding of truth is important in Data Science? - Jan 1, 2018.
Data Science can be used to discover correlations (What phenomena occurred) but cannot be used to establish causality (Why the phenomena occurred).
- How AI Learns What You’re Willing to Pay - Dec 28, 2017.
Why are we all paying different prices? Is it price "personalization" or price "discrimination"? The answer isn't so simple.
- How to Improve Machine Learning Algorithms? Lessons from Andrew Ng, part 2 - Dec 21, 2017.
The second chapter of ML lessons from Ng’s experience. This one will only be talking about Human Level Performance & Avoidable Bias.
- NIPS 2017 Key Points & Summary Notes - Dec 18, 2017.
Third year Ph.D student David Abel, of Brown University, was in attendance at NIP 2017, and he labouriously compiled and formatted a fantastic 43-page set of notes for the rest of us. Get them here.
- Top KDnuggets tweets, Dec 06-12: Top #DataScience and #MachineLearning Methods Used in 2017; Geoff Hinton Capsule Networks – a new way for machines to see - Dec 13, 2017.
Also The first international #beauty contest decided by #AI #algorithm sparked controversy; 4 Common #Data Fallacies That You Need To Know; Using #DeepLearning to Solve Real World Problems; Best Online Masters in #DataScience and #Analytics.
- Machine Ethics and Artificial Moral Agents - Nov 2, 2017.
This article is simply a stream of consciousness on questions and problems I have been thinking and asking myself, and hopefully, it will stimulate some discussion.
Pages: 1 2
- KDnuggets™ News 17:n40, Oct 18: Want to Become a Data Scientist? Read This!; Natural Stupidity is more Dangerous than Artificial Intelligence - Oct 18, 2017.
Want to Become a Data Scientist? Read This Interview First; Natural Stupidity is more Dangerous than Artificial Intelligence; Random Forests(r), Explained; Key Trends and Takeaways from RE-WORK Deep Learning Summit Montreal; An Overview of 3 Popular Courses on Deep Learning
- Data Science and the Imposter Syndrome - Sep 15, 2017.
You are not the only one who wonders how much longer they can get away with pretending to be a data scientist. You are not the only one who has nightmares about being laughed out of your next interview.
- Data Podcast: Gregory Piatetsky-Shapiro, KDnuggets President, a top Big Data Influencer - Aug 22, 2017.
An episode of Data Podcast, featuring Gregory Piatetsky-Shapiro, discussing KDnuggets, trends in Big Data and Machine Learning, Automation of Data Science, Bias in Algorithms and AI, and more.
- O’Reilly NYC AI Conference Highlights: Explainable AI, Vector Representation, Bias, and Future - Aug 21, 2017.
The answer to questions of trust and bias in AI is largely seen in the focus on Explainable AI. Although traditionally viewed as "black boxes", AI and machine learning systems are not ontologically inscrutable.
- Data Science Primer: Basic Concepts for Beginners - Aug 11, 2017.
This collection of concise introductory data science tutorials cover topics including the difference between data mining and statistics, supervised vs. unsupervised learning, and the types of patterns we can mine from data.
- How GDPR Affects Data Science - Jul 17, 2017.
Coming European GDPR directive affects data science practice mainly in 3 areas: limits on data processing and consumer profiling, a “right to an explanation” for automated decision-making, and accountability for bias and discrimination in automated decisions.
- 17 More Must-Know Data Science Interview Questions and Answers - Feb 15, 2017.
17 new must-know Data Science Interview questions and answers include lessons from failure to predict 2016 US Presidential election and Super Bowl LI comeback, understanding bias and variance, why fewer predictors might be better, and how to make a model more robust to outliers.
Pages: 1 2
- The Top Predictive Analytics Pitfalls to Avoid - Jan 23, 2017.
Predictive modelling and machine learning are significantly contributing to business, but they can be very sensitive to data and changes in it, which makes it very important to use proper techniques and avoid pitfalls in building data science models.
- Machine Learning Meets Humans – Insights from HUML 2016 - Jan 6, 2017.
Report from an important IEEE workshop on Human Use of Machine Learning, covering trust, responsibility, the value of explanation, safety of machine learning, discrimination in human vs. machine decision making, and more.
Pages: 1 2
- 4 Reasons Your Machine Learning Model is Wrong (and How to Fix It) - Dec 21, 2016.
This post presents some common scenarios where a seemingly good machine learning model may still be wrong, along with a discussion of how how to evaluate these issues by assessing metrics of bias vs. variance and precision vs. recall.
- 4 Cognitive Bias Key Points Data Scientists Need to Know - Dec 9, 2016.
Cognitive biases are inherently problematic in a variety of fields, including data science. Is this something that can be mitigated? A solid understanding of cognitive biases is the best weapon, which this overview hopes to help provide.
- The Foundations of Algorithmic Bias - Nov 16, 2016.
We might hope that algorithmic decision making would be free of biases. But increasingly, the public is starting to realize that machine learning systems can exhibit these same biases and more. In this post, we look at precisely how that happens.
Pages: 1 2 3
- Top KDnuggets tweets, Aug 03-09: Understanding the Bias-Variance Tradeoff: An Overview - Aug 10, 2016.
Understanding the Bias-Variance Tradeoff: An Overview; Cartoon: Facebook #DataScience experiments and Cats; Bayesian #Machine Learning, Explained; Deep Reinforcement Learning for Keras.
- Understanding the Bias-Variance Tradeoff: An Overview - Aug 8, 2016.
A model's ability to minimize bias and minimize variance are often thought of as 2 opposing ends of a spectrum. Being able to understand these two types of errors are critical to diagnosing model results.
- The Fallacy of Seeing Patterns - Jul 26, 2016.
Analysts are often on the lookout for patterns, often relying on spurious patterns. This post looks at some spurious patterns in univariate, bivariate & multivariate analysis.
- Top KDnuggets tweets, May 4-10: Understanding the Bias-Variance Tradeoff; Python, MachineLearning, & Dueling Languages - May 11, 2016.
Understanding the Bias-Variance Tradeoff; Python, MachineLearning, & Dueling Languages; Why AI development is going to get even faster; Why Implement #MachineLearning Algorithms From Scratch?
- On Why Sequels Are Bad and Red Light Cameras Aren’t As Effective - Feb 3, 2016.
Regression to the mean is a statistical phenomenon whereby extreme observations will tend to decrease (regress) towards the mean on subsequent readings. Regression to the mean is essentially a result of selection bias, learn more about it.
- Data scientists keep forgetting the one rule - Feb 2, 2016.
“Correlation does not imply causation”. Yet data scientists often confuse the two, succumbing to the temptation to over-interpret. And that can lead us to make some really bad decisions from data.
- Data Science and Prejudice – Blessing or Curse ? - Dec 23, 2015.
We examine the deep nature of bias and prejudice and wonder if prejudiced minds and 'good' data scientists coexist in harmony and if they can coexist, does it lead to disruption or disruptive innovation?
- Doubt and Verify: Data Science Power Tools - Jul 3, 2015.
In the end, there is no truth, no ultimate ground truth, no lie-free utterances, as everything is contextual based on incomplete facts and knowledge. All world models are flawed, but Data Science has 2 power tools.
- Interview: Ravi Iyer, Ranker on Dealing with Inherent Bias in Crowdsourcing Data - Apr 8, 2015.
We discuss the challenges of analyzing crowdsourcing data, tools and technologies, competitive landscape, advice, trends, and more.
Pages: 1 2
- Interview: Josh Hemann, Activision on Why the Tolerance for Ambiguity is Vital - Mar 12, 2015.
We discuss handling bias in data, other data quality concerns, advice, desired qualities, and more.
- Top KDnuggets tweets, Mar 12-13: Machine learning explained in 10 pictures; Tutorial: Using Google BigQuery - Mar 14, 2014.
Machine learning explained in 10 pictures. The most important: Bias vs Variance; A Tutorial example: Using Google BigQuery with R; Visualizing Google Analytics Data With R; Exploratory Data Analysis on Udacity: Investigate, Visualize, and Summarize Data Using R.