Researchers and educators have long wrestled with the question of how best to teach their clients be they humans, non-human animals or machines. Here, we examine the role of a single variable, the difficulty of training, on the rate of learning. In many situations we find that there is a sweet spot in which training is neither too easy nor too hard, and where learning progresses most quickly. We derive conditions for this sweet spot for a broad class of learning algorithms in the context of binary classification tasks. For all of these stochastic gradient-descent based learning algorithms, we find that the optimal error rate for training is around 15.87% or, conversely, that the optimal training accuracy is about 85%. We demonstrate the efficacy of this ‘Eighty Five Percent Rule’ for artificial neural networks used in AI and biologically plausible neural networks thought to describe animal learning. Read More
Tag Archives: Accuracy
How Computers See Gender: An Evaluation of Gender Classification in Commercial Facial Analysis and Image Labeling Services
Investigations of facial analysis (FA) technologies—such as facial detection and facial recognition—have been central to discussions about Artificial Intelligence’s (AI) impact on human beings. Research on automatic gender recognition, the classification of gender by FA technologies, has raised potential concerns around issues of racial and gender bias. In this study, we augment past work with empirical data by conducting a systematic analysis of how gender classification and gender labeling in computer vision services operate when faced with gender diversity. We sought to understand how gender is concretely conceptualized and encoded into commercial facial analysis and image labeling technologies available today. We then conducted a two-phrase study: (1) a system analysis of ten commercial FA and image labeling services and (2) an evaluation of five services using a custom dataset of diverse genders using self-labeled Instagram images. Our analysis highlights how gender is codified into both classifiers and data standards. We found that FA services performed consistently worse on transgender individuals and were universally unable to classify non-binary genders. In contrast, image labeling often presented multiple gendered concepts. We also found that user perceptions about gender performance and identity contradict the way gender performance is encoded into the computer vision infrastructure. We discuss our findings from three perspectives of gender identity (self-identity, gender performativity, and demographic identity) and how these perspectives interact across three layers: the classification infrastructure, the third-party applications that make use of that infrastructure, and the individuals who interact with that software. We employ Bowker and Star’s concepts of “torque” and “residuality” to further discuss the social implications of gender classification. We conclude by outlining opportunities for creating more inclusive classification infrastructures and datasets, as well as with implications for policy. Read More
Assuring the Machine Learning Lifecycle: Desiderata, Methods, and Challenges
Machine learning has evolved into an enabling technology for a wide range of highly successful applications. The potential for this success to continue and accelerate has placed machine learning (ML) at the top of research, economic and political agendas. Such unprecedented interest is fueled by a vision of ML applicability extending to healthcare, transportation, defense and other domains of great societal importance. Achieving this vision requires the use of ML in safety-critical applications that demand levels of assurance beyond those needed for current ML applications. Our paper provides a comprehensive survey of the state-of-the-art in the assurance of ML, i.e. in the generation of evidence that ML is sufficiently safe for its intended use. The survey covers the methods capable of providing such evidence at different stages of the machine learning lifecycle, i.e. of the complex, iterative process that starts with the collection of the data used to train an ML component for a system, and ends with the deployment of that component within the system. The paper begins with a systematic presentation of the ML lifecycle and its stages. We then define assurance desiderata for each stage, review existing methods that contribute to achieving these desiderata, and identify open challenges that require further research. Read More
Data in the Life: Authorship Attribution in Lennon-McCartney Songs
The songwriting duo of John Lennon and Paul McCartney, the two founding members of the Beatles, composed some of the most popular and memorable songs of the last century. Despite having authored songs under the joint credit agreement of Lennon-McCartney, it is well-documented that most of their songs or portions of songs were primarily written by exactly one of the two. Furthermore, the authorship of some Lennon-McCartney songs is in dispute, with the recollections of authorship based on previous interviews with Lennon and McCartney in conflict. For Lennon-McCartney songs of known and unknown authorship written and recorded over the period 1962-66, we extracted musical features from each song or song portion. These features consist of the occurrence of melodic notes, chords, melodic note pairs, chord change pairs, and four-note melody contours. We developed a prediction model based on variable screening followed by logistic regression with elastic net regularization. Out-of-sample classification accuracy for songs with known authorship was 76%, with a c-statistic from an ROC analysis of 83.7%. We applied our model to the prediction of songs and song portions with unknown or disputed authorship. Read More
Statistical Significance Tests for Comparing Machine Learning Algorithms
Comparing machine learning methods and selecting a final model is a common operation in applied machine learning.
Models are commonly evaluated using resampling methods like k-fold cross-validation from which mean skill scores are calculated and compared directly. Although simple, this approach can be misleading as it is hard to know whether the difference between mean skill scores is real or the result of a statistical fluke.
Statistical significance tests are designed to address this problem and quantify the likelihood of the samples of skill scores being observed given the assumption that they were drawn from the same distribution. If this assumption, or null hypothesis, is rejected, it suggests that the difference in skill scores is statistically significant.
Although not foolproof, statistical hypothesis testing can improve both your confidence in the interpretation and the presentation of results during model selection. Read More
Rick's Cafe AI 