More Like this

1. Birds of a Feather Flock Together: Satirical News Detection via Language Model Differentiation

   Zhang, Yigeng; Yang, Fan; Zhang, Yifan; Mukherjee, Arjun (October 2020, . International Conference on Social Computing, Behavioral-Cultural Modeling & Prediction and Behavior Representation in Modeling and Simulation)

   null (Ed.)

   Satirical news is regularly shared in modern social media because it is entertaining with smartly embedded humor. However, it can be harmful to society because it can sometimes be mistaken as factual news, due to its deceptive character. We found that in satirical news, the lexical and pragmatical attributes of the context are the key factors in amusing the readers. In this work, we propose a method that differentiates the satirical news and true news. It takes advantage of satirical writing evidence by leveraging the difference between the prediction loss of two language models, one trained on true news and the other on satirical news, when given a new news article. We compute several statistical metrics of language model prediction loss as features, which are then used to conduct downstream classification. The proposed method is computationally effective because the language models capture the language usage differences between satirical news documents and traditional news documents, and are sensitive when applied to documents outside their domains. 

   more » « less
2. Articulating Spatial Statistics and Spatial Optimization Relationships: Expanding the Relevance of Statistics

   https://doi.org/10.3390/stats4040050  

   Griffith, Daniel A. (December 2021, Stats)

   Both historically and in terms of practiced academic organization, the anticipation should be that a flourishing synergistic interface exists between statistics and operations research in general, and between spatial statistics/econometrics and spatial optimization in particular. Unfortunately, for the most part, this expectation is false. The purpose of this paper is to address this existential missing link by focusing on the beneficial contributions of spatial statistics to spatial optimization, via spatial autocorrelation (i.e., dis/similar attribute values tend to cluster together on a map), in order to encourage considerably more future collaboration and interaction between contributors to their two parent bodies of knowledge. The key basic statistical concept in this pursuit is the median in its bivariate form, with special reference to the global and to sets of regional spatial medians. One-dimensional examples illustrate situations that the narrative then extends to two-dimensional illustrations, which, in turn, connects these treatments to the spatial statistics centrography theme. Because of computational time constraints (reported results include some for timing experiments), the summarized analysis restricts attention to problems involving one global and two or three regional spatial medians. The fundamental and foundational spatial, statistical, conceptual tool employed here is spatial autocorrelation: geographically informed sampling designs—which acknowledge a non-random mixture of geographic demand weight values that manifests itself as local, homogeneous, spatial clusters of these values—can help spatial optimization techniques determine the spatial optima, at least for location-allocation problems. A valuable discovery by this study is that existing but ignored spatial autocorrelation latent in georeferenced demand point weights undermines spatial optimization algorithms. All in all, this paper should help initiate a dissipation of the existing isolation between statistics and operations research, hopefully inspiring substantially more collaborative work by their professionals in the future. 

   more » « less
3. Spectrally Transformed Kernel Regression

   Zhai, Runtian; Pukdee, Rattana; Jin, Roger; Balcan, Maria-Florina; Ravikumar, Pradeep (May 2024, International Conference on Learning Representations (ICLR), 2024)

   Unlabeled data is a key component of modern machine learning. In general, the role of unlabeled data is to impose a form of smoothness, usually from the similarity information encoded in a base kernel, such as the ε-neighbor kernel or the adjacency matrix of a graph. This work revisits the classical idea of spectrally transformed kernel regression (STKR), and provides a new class of general and scalable STKR estimators able to leverage unlabeled data. Intuitively, via spectral transformation, STKR exploits the data distribution for which unlabeled data can provide additional information. First, we show that STKR is a principled and general approach, by characterizing a universal type of “target smoothness”, and proving that any sufficiently smooth function can be learned by STKR. Second, we provide scalable STKR implementations for the inductive setting and a general transformation function, while prior work is mostly limited to the transductive setting. Third, we derive statistical guarantees for two scenarios: STKR with a known polynomial transformation, and STKR with kernel PCA when the transformation is unknown. Overall, we believe that this work helps deepen our understanding of how to work with unlabeled data, and its generality makes it easier to inspire new methods. 

   more » « less
4. Principled Statistical Inference in Data Science

   https://doi.org/10.1142/9781786345400_0002  

   Kuffner, Todd A.; Young, G.A. (June 2018, Statistical Data Science)

   We discuss the challenges of principled statistical inference in modern data science. Conditionality principles are argued as key to achieving valid statistical inference, in particular when this is performed after selecting a model from sample data itself. 

   more » « less
5. Comparing Text Representations: A Theory-Driven Approach

   https://doi.org/10.18653/v1/2021.emnlp-main.449  

   Yauney, Gregory; Mimno, David (January 2021, Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing (EMNLP))

   Much of the progress in contemporary NLP has come from learning representations, such as masked language model (MLM) contextual embeddings, that turn challenging problems into simple classification tasks. But how do we quantify and explain this effect? We adapt general tools from computational learning theory to fit the specific characteristics of text datasets and present a method to evaluate the compatibility between representations and tasks. Even though many tasks can be easily solved with simple bag-of-words (BOW) representations, BOW does poorly on hard natural language inference tasks. For one such task we find that BOW cannot distinguish between real and randomized labelings, while pre-trained MLM representations show 72x greater distinction between real and random labelings than BOW. This method provides a calibrated, quantitative measure of the difficulty of a classification-based NLP task, enabling comparisons between representations without requiring empirical evaluations that may be sensitive to initializations and hyperparameters. The method provides a fresh perspective on the patterns in a dataset and the alignment of those patterns with specific labels. 

   more » « less