More Like this

1. The Bias Amplification Paradox in Text-to-Image Generation

   Seshadri, Preethi; Singh, Sameer; Elazar, Yanai (June 2024, Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)))

   Bias amplification is a phenomenon in which models exacerbate biases or stereotypes present in the training data. In this paper, we study bias amplification in the text-to-image domain using Stable Diffusion by comparing gender ratios in training vs. generated images. We find that the model appears to amplify gender-occupation biases found in the training data (LAION) considerably. However, we discover that amplification can be largely attributed to discrepancies between training captions and model prompts. For example, an inherent difference is that captions from the training data often contain explicit gender information while our prompts do not, which leads to a distribution shift and consequently inflates bias measures. Once we account for distributional differences between texts used for training and generation when evaluating amplification, we observe that amplification decreases drastically. Our findings illustrate the challenges of comparing biases in models and their training data, as well as evaluation more broadly, and highlight how confounding factors can impact analyses. 

   more » « less
2. Feature-Wise Bias Amplification

   Leino, Klas; Fredrikson, Matt; Black, Emily; Sen, Shayak; Datta, Anupam (January 2019, International Conference on Learning Representations (ICLR))

   We study the phenomenon of bias amplification in classifiers, wherein a machine learning model learns to predict classes with a greater disparity than the underlying ground truth. We demonstrate that bias amplification can arise via inductive bias in gradient descent methods resulting in overestimation of importance of moderately-predictive weak'' features if insufficient training data is available. This overestimation gives rise to feature-wise bias amplification -- a previously unreported form of bias that can be traced back to the features of a trained model. Through analysis and experiments, we show that the while some bias cannot be mitigated without sacrificing accuracy, feature-wise bias amplification can be mitigated through targeted feature selection. We present two new feature selection algorithms for mitigating bias amplification in linear models, and show how they can be adapted to convolutional neural networks efficiently. Our experiments on synthetic and real data demonstrate that these algorithms consistently lead to reduced bias without harming accuracy, in some cases eliminating predictive bias altogether while providing modest gains in accuracy. 

   more » « less
3. Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints

   https://doi.org/10.18653/v1/D17-1323  

   Zhao, Jieyu; Wang, Tianlu; Yatskar, Mark; Ordonez, Vicente; Chang, Kai-Wei (September 2017, Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing)

   Language is increasingly being used to define rich visual recognition problems with supporting image collections sourced from the web. Structured prediction models are used in these tasks to take advantage of correlations between co-occurring labels and visual input but risk inadvertently encoding social biases found in web corpora. In this work, we study data and models associated with multilabel object classification and visual semantic role labeling. We find that (a) datasets for these tasks contain significant gender bias and (b) models trained on these datasets further amplify existing bias. For example, the activity cooking is over 33% more likely to involve females than males in a training set, and a trained model further amplifies the disparity to 68% at test time. We propose to inject corpus-level constraints for calibrating existing structured prediction models and design an algorithm based on Lagrangian relaxation for collective inference. Our method results in almost no performance loss for the underlying recognition task but decreases the magnitude of bias amplification by 47.5% and 40.5% for multilabel classification and visual semantic role labeling, respectively 

   more » « less
4. Development of planar P-type point contact germanium detectors for low-mass dark matter searches

   https://doi.org/10.1140/epjc/s10052-022-10162-x  

   Wei, W.-Z.; Mei, H.; Kooi, K.; Mei, D.-M.; Liu, J.; Li, J.-C.; Panth, R.; Wang, G.-J. (March 2022, The European Physical Journal C)

   Abstract The detection of low-energy deposition in the range of sub-eV through ionization using germanium (Ge) with a bandgap of $$\sim $$ ∼ 0.7 eV requires internal amplification of the charge signal. This can be achieved through high electric field that accelerates charge carriers, which can then generate more charge carriers. The minimum electric field required to generate internal charge amplification is derived for different temperatures. We report the development of a planar point contact Ge detector in terms of its fabrication and the measurements of its leakage current and capacitance as a function of applied bias voltage. With the determination of the measured depletion voltage, the field distribution is calculated using GeFiCa, which predicts that the required electric field for internal charge amplification can be achieved in proximity to the point contact. The energy response to an Am-241 source is characterized and discussed. We conclude that such a detector with internal charge amplification can be used to search for low-mass dark matter. 

   more » « less
5. Sample-Optimal Identity Testing with High Probability

   Diakonikolas, Ilias; Gouleakis, Themis; Peebles, John; Price, Eric (July 2018, 45th International Colloquium on Automata, Languages, and Automata)

   We study the problem of testing identity against a given distribution with a focus on the high confidence regime. More precisely, given samples from an unknown distribution p over n elements, an explicitly given distribution q, and parameters 0< epsilon, delta < 1, we wish to distinguish, with probability at least 1-delta, whether the distributions are identical versus epsilon-far in total variation distance. Most prior work focused on the case that delta = Omega(1), for which the sample complexity of identity testing is known to be Theta(sqrt{n}/epsilon^2). Given such an algorithm, one can achieve arbitrarily small values of delta via black-box amplification, which multiplies the required number of samples by Theta(log(1/delta)). We show that black-box amplification is suboptimal for any delta = o(1), and give a new identity tester that achieves the optimal sample complexity. Our new upper and lower bounds show that the optimal sample complexity of identity testing is Theta((1/epsilon^2) (sqrt{n log(1/delta)} + log(1/delta))) for any n, epsilon, and delta. For the special case of uniformity testing, where the given distribution is the uniform distribution U_n over the domain, our new tester is surprisingly simple: to test whether p = U_n versus d_{TV} (p, U_n) >= epsilon, we simply threshold d_{TV}({p^}, U_n), where {p^} is the empirical probability distribution. The fact that this simple "plug-in" estimator is sample-optimal is surprising, even in the constant delta case. Indeed, it was believed that such a tester would not attain sublinear sample complexity even for constant values of epsilon and delta. An important contribution of this work lies in the analysis techniques that we introduce in this context. First, we exploit an underlying strong convexity property to bound from below the expectation gap in the completeness and soundness cases. Second, we give a new, fast method for obtaining provably correct empirical estimates of the true worst-case failure probability for a broad class of uniformity testing statistics over all possible input distributions - including all previously studied statistics for this problem. We believe that our novel analysis techniques will be useful for other distribution testing problems as well. 

   more » « less