<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>AnthropocentricAI</title>
    <description></description>
    <link>https://anthropocentricai.github.io/</link>
    <atom:link href="https://anthropocentricai.github.io/feed.xml" rel="self" type="application/rss+xml"/>
    <pubDate>Fri, 17 May 2019 15:09:01 +0000</pubDate>
    <lastBuildDate>Fri, 17 May 2019 15:09:01 +0000</lastBuildDate>
    <generator>Jekyll v3.8.5</generator>
    
      <item>
        <title>Circle CI Setup</title>
        <description>
CircleCI integration and configuration file setup.

</description>
        <pubDate>Thu, 25 Oct 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/10/25/circleCI/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/10/25/circleCI/</guid>
        
        <category>CI</category>
        
        <category>continuous-integration</category>
        
        <category>circle</category>
        
        
        <category>software-engineering</category>
        
      </item>
    
      <item>
        <title>Travis CI</title>
        <description>
Setting up continuous integration with &lt;a href=&quot;https://travis-ci.org&quot;&gt;Travis&lt;/a&gt;.

</description>
        <pubDate>Fri, 28 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/28/travis/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/28/travis/</guid>
        
        <category>Travis</category>
        
        <category>continuous-integration</category>
        
        <category>CI</category>
        
        <category>Codecov</category>
        
        
        <category>software-engineering</category>
        
      </item>
    
      <item>
        <title>Evaluating Fairness Metrics in the Presence of Dataset Bias</title>
        <description>&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#abstract&quot; id=&quot;markdown-toc-abstract&quot;&gt;Abstract&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&quot;abstract&quot;&gt;Abstract&lt;/h2&gt;
Data-driven algorithms play a large role in decision making across a variety of industries. Increasingly, these algorithms are being used to make decisions that have significant ramifications for people’s social and economic well-being, e.g. in sentencing, loan approval, and policing. Amid the proliferation of such systems there is a growing concern about their potential discriminatory impact. In particular, machine learning systems which are trained on biased data have the potential to learn and perpetuate those biases. A central challenge for practitioners is thus to determine whether their models display discriminatory bias. Here we present a case study in which we frame the issue of bias detection as a causal inference problem with observational data. We enumerate two main causes of bias, sampling bias and label bias, and we investigate the abilities of six different fairness metrics to detect each bias type. Based on these investigations, we propose a set of best practice guidelines to select the fairness metric that is most likely to detect bias if it is present. Additionally, we aim to identify the conditions in which certain fairness metrics may fail to detect bias and instead give practitioners a false belief that their biased model is making fair decisions.

</description>
        <pubDate>Wed, 26 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/26/hinnefeld2018evaluating/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/26/hinnefeld2018evaluating/</guid>
        
        <category>fairness</category>
        
        
        <category>paper</category>
        
      </item>
    
      <item>
        <title>Python Linting Software</title>
        <description>
Linting software for python.

</description>
        <pubDate>Mon, 24 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/24/linting/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/24/linting/</guid>
        
        <category>code-quality</category>
        
        <category>linting</category>
        
        
        <category>software-engineering</category>
        
      </item>
    
      <item>
        <title>Open Source Licensing</title>
        <description>
Open source licensing comparison report.

</description>
        <pubDate>Thu, 20 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/20/license-report/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/20/license-report/</guid>
        
        <category>licence</category>
        
        
        <category>software-engineering</category>
        
      </item>
    
      <item>
        <title>Foolbox: A python toolbox to benchmark the robustness of machine learning models</title>
        <description>&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#abstract&quot; id=&quot;markdown-toc-abstract&quot;&gt;Abstract&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&quot;abstract&quot;&gt;Abstract&lt;/h2&gt;
Even today’s most advanced machine learning models are easily fooled by almost imperceptible perturbations of their inputs. Foolbox is a new Python package to generate such adversarial perturbations and to quantify and compare the robustness of machine learning models. It is build around the idea that the most comparable robustness measure is the minimum perturbation needed to craft an adversarial example. To this end, Foolbox provides reference implementations of most published adversarial attack methods alongside some new ones, all of which perform internal hyper-parameter tuning to find the minimum adversarial perturbation. Additionally, Foolbox interfaces with most popular deep learning frameworks such as PyTorch, Keras, TensorFlow, Theano and MXNet and allows different adversarial criteria such as targeted misclassification and top-k misclassification as well as different distance measures. The code is licensed under the MIT license and is openly available at &lt;a href=&quot;https://github.com/bethgelab/foolbox&quot;&gt;this https URL&lt;/a&gt;. The most up-to-date documentation can be found at &lt;a href=&quot;http://foolbox.readthedocs.io/&quot;&gt;this http URL&lt;/a&gt;.

</description>
        <pubDate>Thu, 20 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/20/rauber1707foolbox/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/20/rauber1707foolbox/</guid>
        
        <category>robustness</category>
        
        <category>accountability</category>
        
        
        <category>paper</category>
        
      </item>
    
      <item>
        <title>Fairness-Aware Classifier with Prejudice Remover Regularizer</title>
        <description>&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#abstract&quot; id=&quot;markdown-toc-abstract&quot;&gt;Abstract&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&quot;abstract&quot;&gt;Abstract&lt;/h2&gt;
With the spread of data mining technologies and the accumulation of social data, such technologies and data are being used for determinations that seriously affect individuals’ lives. For example, credit scoring is frequently determined based on the records of past credit data together with statistical prediction techniques. Needless to say, such determinations must be nondiscriminatory and fair in sensitive features, such as race, gender, religion, and so on. Several researchers have recently begun to attempt the development of analysis techniques that are aware of social fairness or discrimination. They have shown that simply avoiding the use of sensitive features is insufficient for eliminating biases in determinations, due to the indirect influence of sensitive information. In this paper, we first discuss three causes of unfairness in machine learning. We then propose a regularization approach that is applicable to any prediction algorithm with probabilistic discriminative models. We further apply this approach to logistic regression and empirically show its effectiveness and efficiency.

</description>
        <pubDate>Sun, 02 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/02/kamishima2012fairness/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/02/kamishima2012fairness/</guid>
        
        <category>fairness</category>
        
        
        <category>paper</category>
        
      </item>
    
      <item>
        <title>A comparative study of fairness-enhancing interventions in machine learning</title>
        <description>&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#abstract&quot; id=&quot;markdown-toc-abstract&quot;&gt;Abstract&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&quot;abstract&quot;&gt;Abstract&lt;/h2&gt;
Computers are increasingly used to make decisions that have significant impact in people’s lives. Often, these predictions can affect different population subgroups disproportionately. As a result, the issue of fairness has received much recent interest, and a number of fairness-enhanced classifiers and predictors have appeared in the literature. This paper seeks to study the following questions: how do these different techniques fundamentally compare to one another, and what accounts for the differences? Specifically, we seek to bring attention to many under-appreciated aspects of such fairness-enhancing interventions. Concretely, we present the results of an open benchmark we have developed that lets us compare a number of different algorithms under a variety of fairness measures, and a large number of existing datasets. We find that although different algorithms tend to prefer specific formulations of fairness preservations, many of these measures strongly correlate with one another. In addition, we find that fairness-preserving algorithms tend to be sensitive to fluctuations in dataset composition (simulated in our benchmark by varying training-test splits), indicating that fairness interventions might be more brittle than previously thought.

</description>
        <pubDate>Sun, 02 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/02/friedler2018comparative/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/02/friedler2018comparative/</guid>
        
        <category>fairness</category>
        
        
        <category>paper</category>
        
      </item>
    
      <item>
        <title>Certifying and removing disparate impact</title>
        <description>&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#abstract&quot; id=&quot;markdown-toc-abstract&quot;&gt;Abstract&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&quot;abstract&quot;&gt;Abstract&lt;/h2&gt;
What does it mean for an algorithm to be biased? In U.S. law, unintentional bias is encoded via disparate impact, which occurs when a selection process has widely different outcomes for different groups, even as it appears to be neutral. This legal determination hinges on a definition of a protected class (ethnicity, gender, religious practice) and an explicit description of the process. 
When the process is implemented using computers, determining disparate impact (and hence bias) is harder. It might not be possible to disclose the process. In addition, even if the process is open, it might be hard to elucidate in a legal setting how the algorithm makes its decisions. Instead of requiring access to the algorithm, we propose making inferences based on the data the algorithm uses. 
We make four contributions to this problem. First, we link the legal notion of disparate impact to a measure of classification accuracy that while known, has received relatively little attention. Second, we propose a test for disparate impact based on analyzing the information leakage of the protected class from the other data attributes. Third, we describe methods by which data might be made unbiased. Finally, we present empirical evidence supporting the effectiveness of our test for disparate impact and our approach for both masking bias and preserving relevant information in the data. Interestingly, our approach resembles some actual selection practices that have recently received legal scrutiny.

</description>
        <pubDate>Sun, 02 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/02/feldman2015certifying/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/02/feldman2015certifying/</guid>
        
        <category>fairness</category>
        
        
        <category>paper</category>
        
      </item>
    
      <item>
        <title>Auditing black-box models for indirect influence</title>
        <description>&lt;ul id=&quot;markdown-toc&quot;&gt;
  &lt;li&gt;&lt;a href=&quot;#abstract&quot; id=&quot;markdown-toc-abstract&quot;&gt;Abstract&lt;/a&gt;&lt;/li&gt;
&lt;/ul&gt;

&lt;h2 id=&quot;abstract&quot;&gt;Abstract&lt;/h2&gt;
Data-trained predictive models see widespread use, but for the most part they are used as black boxes which output a prediction or score. It is therefore hard to acquire a deeper understanding of model behavior, and in particular how different features influence the model prediction. This is important when interpreting the behavior of complex models, or asserting that certain problematic attributes (like race or gender) are not unduly influencing decisions. 
In this paper, we present a technique for auditing black-box models, which lets us study the extent to which existing models take advantage of particular features in the dataset, without knowing how the models work. Our work focuses on the problem of indirect influence: how some features might indirectly influence outcomes via other, related features. As a result, we can find attribute influences even in cases where, upon further direct examination of the model, the attribute is not referred to by the model at all. 
Our approach does not require the black-box model to be retrained. This is important if (for example) the model is only accessible via an API, and contrasts our work with other methods that investigate feature influence like feature selection. We present experimental evidence for the effectiveness of our procedure using a variety of publicly available datasets and models. We also validate our procedure using techniques from interpretable learning and feature selection, as well as against other black-box auditing procedures.

</description>
        <pubDate>Sun, 02 Sep 2018 00:00:00 +0000</pubDate>
        <link>https://anthropocentricai.github.io/blog/2018/09/02/adler2018auditing/</link>
        <guid isPermaLink="true">https://anthropocentricai.github.io/blog/2018/09/02/adler2018auditing/</guid>
        
        <category>fairness</category>
        
        
        <category>paper</category>
        
      </item>
    
  </channel>
</rss>
