A significant impediment to progress in research on bias in machine learning (ML) is the availability of relevant datasets. This situation is unlikely to change much given the sensitivity of such data. For this reason, there is a role for synthetic data in this research. In this short paper, we present one such family of synthetic data sets. We provide an overview of the data, describe how the level of bias can be varied, and present a simple example of an experiment on the data.
While state-of-the-art NLP models have been achieving the excellent performance of a wide range of tasks in recent years, important questions are being raised about their robustness and their underlying sensitivity to systematic biases that may exist in their training and test data. Such issues come to be manifest in performance problems when faced with out-of-distribution data in the field. One recent solution has been to use counterfactually augmented datasets in order to reduce any reliance on spurious patterns that may exist in the original data. Producing high-quality augmented data can be costly and time-consuming as it usually needs to involve human feedback and crowdsourcing efforts. In this work, we propose an alternative by describing and evaluating an approach to automatically generating counterfactual data for data augmentation and explanation. A comprehensive evaluation on several different datasets and using a variety of state-of-the-art benchmarks demonstrate how our approach can achieve significant improvements in model performance when compared to models training on the original data and even when compared to models trained with the benefit of human-generated augmented data.
Algorithmic Bias can be due to bias in the training data or issues with the algorithm itself. These algorithmic issues typically relate to problems with model capacity and regularisation. This underestimation bias may arise because the model has been optimised for good generalisation accuracy without any explicit consideration of bias or fairness. In a sense, we should not be surprised that a model might be biased when it hasn't been "asked" not to be. In this paper, we consider including bias (underestimation) as an additional criterion in model training. We present a multi-objective optimisation strategy using Pareto Simulated Annealing that optimise for both balanced accuracy and underestimation. We demonstrate the effectiveness of this strategy on one synthetic and two real-world datasets.
It is fair to say that many of the prominent examples of bias in Machine Learning (ML) arise from bias that is there in the training data. In fact, some would argue that supervised ML algorithms cannot be biased, they reflect the data on which they are trained. In this paper we demonstrate how ML algorithms can misrepresent the training data through underestimation. We show how irreducible error, regularization and feature and class imbalance can contribute to this underestimation. The paper concludes with a demonstration of how the careful management of synthetic counterfactuals can ameliorate the impact of this underestimation bias.
It is well understood that Dynamic Time Warping (DTW) is effective in revealing similarities between time series that do not align perfectly. In this paper, we illustrate this on spectroscopy time-series data. We show that DTW is effective in improving accuracy on a regression task when only a single wavelength is considered. When combined with k-Nearest Neighbour, DTW has the added advantage that it can reveal similarities and differences between samples at the level of the time-series. However, in the problem, we consider here data is available across a spectrum of wavelengths. If aggregate statistics (means, variances) are used across many wavelengths the benefits of DTW are no longer apparent. We present this as another example of a situation where big data trumps sophisticated models in Machine Learning.
In this paper we conduct an analysis of Moodle activity data focused on identifying early predictors of good student performance. The analysis shows that three relevant hypotheses are largely supported by the data. These hypotheses are: early submission is a good sign, a high level of activity is predictive of good results and evening activity is even better than daytime activity. We highlight some pathological examples where high levels of activity correlates with bad results.
Twitter is often the most up-to-date source for finding and tracking breaking news stories. Therefore, there is considerable interest in developing filters for tweet streams in order to track and summarize stories. This is a non-trivial text analytics task as tweets are short, and standard retrieval methods often fail as stories evolve over time. In this paper we examine the effectiveness of adaptive mechanisms for tracking and summarizing breaking news stories. We evaluate the effectiveness of these mechanisms on a number of recent news events for which manually curated timelines are available. Assessments based on ROUGE metrics indicate that an adaptive approaches are best suited for tracking evolving stories on Twitter.
Collaborations such as Wikipedia are a key part of the value of the modern Internet. At the same time there is concern that these collaborations are threatened by high levels of member turnover. In this paper we borrow ideas from topic analysis to editor activity on Wikipedia over time into a latent space that offers an insight into the evolving patterns of editor behavior. This latent space representation reveals a number of different categories of editor (e.g. content experts, social networkers) and we show that it does provide a signal that predicts an editor's departure from the community. We also show that long term editors gradually diversify their participation by shifting edit preference from one or two namespaces to multiple namespaces and experience relatively soft evolution in their editor profiles, while short term editors generally distribute their contribution randomly among the namespaces and experience considerably fluctuated evolution in their editor profiles.
Topic modeling refers to the task of discovering the underlying thematic structure in a text corpus, where the output is commonly presented as a report of the top terms appearing in each topic. Despite the diversity of topic modeling algorithms that have been proposed, a common challenge in successfully applying these techniques is the selection of an appropriate number of topics for a given corpus. Choosing too few topics will produce results that are overly broad, while choosing too many will result in the "over-clustering" of a corpus into many small, highly-similar topics. In this paper, we propose a term-centric stability analysis strategy to address this issue, the idea being that a model with an appropriate number of topics will be more robust to perturbations in the data. Using a topic modeling approach based on matrix factorization, evaluations performed on a range of corpora show that this strategy can successfully guide the model selection process.