Search for: All records

Word vector embeddings have been shown to contain and amplify biases in the data they are extracted from. Consequently, many techniques have been proposed to identify, mitigate, and attenuate these biases in word representations. In this paper, we utilize interactive visualization to increase the interpretability and accessibility of a collection of state-of-the-art debiasing techniques. To aid this, we present the Visualization of Embedding Representations for deBiasing (“VERB”) system, an open-source web-based visualization tool that helps users gain a technical understanding and visual intuition of the inner workings of debiasing techniques, with a focus on their geometric properties. In particular, VERB offers easy-to-follow examples that explore the effects of these debiasing techniques on the geometry of high-dimensional word vectors. To help understand how various debiasing techniques change the underlying geometry, VERB decomposes each technique into interpretable sequences of primitive transformations and highlights their effect on the word vectors using dimensionality reduction and interactive visual exploration. VERB is designed to target natural language processing (NLP) practitioners who are designing decision-making systems on top of word embeddings, and also researchers working with the fairness and ethics of machine learning systems in NLP. It can also serve as a visual medium for education, which helps an NLP novice understand and mitigate biases in word embeddings. 

Word vector embeddings have been shown to contain and amplify biases in the data they are extracted from. Consequently, many techniques have been proposed to identify, mitigate, and attenuate these biases in word representations. In this paper, we utilize interactive visualization to increase the interpretability and accessibility of a collection of state-of-the-art debiasing techniques. To aid this, we present the Visualization of Embedding Representations for deBiasing (“VERB”) system, an open-source web-based visualization tool that helps users gain a technical understanding and visual intuition of the inner workings of debiasing techniques, with a focus on their geometric properties. In particular, VERB offers easy-to-follow examples that explore the effects of these debiasing techniques on the geometry of high-dimensional word vectors. To help understand how various debiasing techniques change the underlying geometry, VERB decomposes each technique into interpretable sequences of primitive transformations and highlights their effect on the word vectors using dimensionality reduction and interactive visual exploration. VERB is designed to target natural language processing (NLP) practitioners who are designing decision-making systems on top of word embeddings, and also researchers working with the fairness and ethics of machine learning systems in NLP. It can also serve as a visual medium for education, which helps an NLP novice understand and mitigate biases in word embeddings. 

Public opinion surveys constitute a widespread, powerful tool to study peoples’ attitudes and behaviors from comparative perspectives. However, even global surveys can have limited geographic and temporal coverage, which can hinder the production of comprehensive knowledge. To expand the scope of comparison, social scientists turn to ex-post harmonization of variables from datasets that cover similar topics but in different populations and/or at different times. These harmonized datasets can be analyzed as a single source and accessed through various data portals. However, the Survey Data Recycling (SDR) research project has identified three challenges faced by social scientists when using data portals: the lack of capability to explore data in-depth or query data based on customized needs, the difficulty in efficiently identifying related data for studies, and the incapability to evaluate theoretical models using sliced data. To address these issues, the SDR research project has developed the SDR Querier, which is applied to the harmonized SDR database. The SDR Querier includes a BERT-based model that allows for customized data queries through research questions or keywords (Query-by-Question), a visual design that helps users determine the availability of harmonized data for a given research question (Query-by-Condition), and the ability to reveal the underlying relational patterns among substantive and methodological variables in the database (Query-by-Relation), aiding in the rigorous evaluation or improvement of regression models. Case studies with multiple social scientists have demonstrated the usefulness and effectiveness of the SDR Querier in addressing daily challenges. 

Single-molecule force spectroscopy is a powerful tool for the quantitative investigation of the biophysics, polymer physics and mechanochemistry of individual polymer strands. One limitation of this technique is that the attachment between the tip of the atomic force microscope and the covalent or noncovalent analyte in a given pull is typically not strong enough to sustain the force at which the event of interest occurs, which makes the experiments time-consuming and inhibits throughput. Here we report a polyelectrolyte handle for single-molecule force spectroscopy that offers a combination of high (several hundred pN) attachment forces, good (~4%) success in obtaining a high-force (>200 pN) attachment, a non-fouling detachment process that allows for repetition, and specific attachment locations along the polymer analyte. 

Polymers with low ceiling temperatures (T_c) are highly desirable as they can depolymerize under mild conditions, but they typically suffer from demanding synthetic conditions and poor stability. We envision that this challenge can be addressed by developing high-T_cpolymers that can be converted into low-T_cpolymers on demand. Here, we demonstrate the mechanochemical generation of a low-T_cpolymer, poly(2,5-dihydrofuran) (PDHF), from an unsaturated polyether that contains cyclobutane-fused THF in each repeat unit. Upon mechanically induced cycloreversion of cyclobutane, each repeat unit generates three repeat units of PDHF. The resulting PDHF completely depolymerizes into 2,5-dihydrofuran in the presence of a ruthenium catalyst. The mechanochemical generation of the otherwise difficult-to-synthesize PDHF highlights the power of polymer mechanochemistry in accessing elusive structures. The concept of mechanochemically regulating theT_cof polymers can be applied to develop next-generation sustainable plastics.