This paper proposes approaches to automatically create a large number of new bilingual dictionaries for low-resource languages, especially resource-poor and endangered languages, from a single input bilingual dictionary. Our algorithms produce translations of words in a source language to plentiful target languages using available Wordnets and a machine translator (MT). Since our approaches rely on just one input dictionary, available Wordnets and an MT, they are applicable to any bilingual dictionary as long as one of the two languages is English or has a Wordnet linked to the Princeton Wordnet. Starting with 5 available bilingual dictionaries, we create 48 new bilingual dictionaries. Of these, 30 pairs of languages are not supported by the popular MTs: Google and Bing.

Large transformer-based pre-trained language models have achieved impressive performance on a variety of knowledge-intensive tasks and can capture factual knowledge in their parameters. We argue that storing large amounts of knowledge in the model parameters is sub-optimal given the ever-growing amounts of knowledge and resource requirements. We posit that a more efficient alternative is to provide explicit access to contextually relevant structured knowledge to the model and train it to use that knowledge. We present LM-CORE -- a general framework to achieve this -- that allows \textit{decoupling} of the language model training from the external knowledge source and allows the latter to be updated without affecting the already trained model. Experimental results show that LM-CORE, having access to external knowledge, achieves significant and robust outperformance over state-of-the-art knowledge-enhanced language models on knowledge probing tasks; can effectively handle knowledge updates; and performs well on two downstream tasks. We also present a thorough error analysis highlighting the successes and failures of LM-CORE.

Sr. Project Coordinator, IT at Salesforce Role requiring ‘No experience data provided’ months of experience in Indianapolis To get the

Polarization and echo chambers are often studied in the context of explicitly political events such as elections, and little scholarship has examined the mixing of political groups in non-political contexts. A major obstacle to studying political polarization in non-political contexts is that political leaning (i.e., left vs right orientation) is often unknown. Nonetheless, political leaning is known to correlate (sometimes quite strongly) with many lifestyle choices leading to stereotypes such as the "latte-drinking liberal." We develop a machine learning classifier to infer political leaning from non-political text and, optionally, the accounts a user follows on social media. We use Voter Advice Application results shared on Twitter as our groundtruth and train and test our classifier on a Twitter dataset comprising the 3,200 most recent tweets of each user after removing any tweets with political text. We correctly classify the political leaning of most users (F1 scores range from 0.70 to 0.85 depending on coverage). We find no relationship between the level of political activity and our classification results. We apply our classifier to a case study of news sharing in the UK and discover that, in general, the sharing of political news exhibits a distinctive left-right divide while sports news does not.

With the motive of training all the parameters of a neural network, we study why and when one can achieve this by iteratively creating, training, and combining randomly selected subnetworks. Such scenarios have either implicitly or explicitly emerged in the recent literature: see e.g., the Dropout family of regularization techniques, or some distributed ML training protocols that reduce communication/computation complexities, such as the Independent Subnet Training protocol. While these methods are studied empirically and utilized in practice, they often enjoy partial or no theoretical support, especially when applied on neural network-based objectives. In this manuscript, our focus is on overparameterized single hidden layer neural networks with ReLU activations in the lazy training regime. By carefully analyzing $i)$ the subnetworks' neural tangent kernel, $ii)$ the surrogate functions' gradient, and $iii)$ how we sample and combine the surrogate functions, we prove linear convergence rate of the training error -- up to a neighborhood around the optimal point -- for an overparameterized single-hidden layer perceptron with a regression loss. Our analysis reveals a dependency of the size of the neighborhood around the optimal point on the number of surrogate models and the number of local training steps for each selected subnetwork. Moreover, the considered framework generalizes and provides new insights on dropout training, multi-sample dropout training, as well as Independent Subnet Training; for each case, we provide convergence results as corollaries of our main theorem.

We investigate how to modify executable files to deceive malware classification systems. This work's main contribution is a methodology to inject bytes across a malware file randomly and use it both as an attack to decrease classification accuracy but also as a defensive method, augmenting the data available for training. It respects the operating system file format to make sure the malware will still execute after our injection and will not change its behavior. We reproduced five state-of-the-art malware classification approaches to evaluate our injection scheme: one based on GIST+KNN, three CNN variations and one Gated CNN. We performed our experiments on a public dataset with 9,339 malware samples from 25 different families. Our results show that a mere increase of 7% in the malware size causes an accuracy drop between 25% and 40% for malware family classification. They show that a automatic malware classification system may not be as trustworthy as initially reported in the literature. We also evaluate using modified malwares alongside the original ones to increase networks robustness against mentioned attacks. Results show that a combination of reordering malware sections and injecting random data can improve overall performance of the classification. Code available at https://github.com/adeilsonsilva/malware-injection.

Spoken language recognition (SLR) refers to the automatic process used to determine the language present in a speech sample. SLR is an important task in its own right, for example, as a tool to analyze or categorize large amounts of multi-lingual data. Further, it is also an essential tool for selecting downstream applications in a work flow, for example, to chose appropriate speech recognition or machine translation models. SLR systems are usually composed of two stages, one where an embedding representing the audio sample is extracted and a second one which computes the final scores for each language. In this work, we approach the SLR task as a detection problem and implement the second stage as a probabilistic linear discriminant analysis (PLDA) model. We show that discriminative training of the PLDA parameters gives large gains with respect to the usual generative training. Further, we propose a novel hierarchical approach where two PLDA models are trained, one to generate scores for clusters of highly-related languages and a second one to generate scores conditional to each cluster. The final language detection scores are computed as a combination of these two sets of scores. The complete model is trained discriminatively to optimize a cross-entropy objective. We show that this hierarchical approach consistently outperforms the non-hierarchical one for detection of highly related languages, in many cases by large margins. We train our systems on a collection of datasets including over 100 languages, and test them both on matched and mismatched conditions, showing that the gains are robust to condition mismatch.

The graph represents a network of 1,513 Twitter users whose tweets in the requested range contained "iiot ai", or who were replied to or mentioned in those tweets. The network was obtained from the NodeXL Graph Server on Friday, 29 July 2022 at 10:40 UTC. The requested start date was Friday, 29 July 2022 at 00:01 UTC and the maximum number of tweets (going backward in time) was 7,500. The tweets in the network were tweeted over the 2-day, 17-hour, 57-minute period from Tuesday, 26 July 2022 at 06:02 UTC to Friday, 29 July 2022 at 00:00 UTC. Additional tweets that were mentioned in this data set were also collected from prior time periods.

Interventions to increase active commuting have been recommended as a method to increase population physical activity, but evidence is mixed. Social norms related to travel behaviour may influence the uptake of active commuting interventions but are rarely considered in their design and evaluation. In this study we develop an agent-based model that incorporates social norms related to travel behaviour and demonstrate the utility of this through implementing car-free Wednesdays. A synthetic population of Waltham Forest, London, UK was generated using a microsimulation approach with data from the UK Census 2011 and UK HLS datasets. An agent-based model was created using this synthetic population which modelled how the actions of peers and neighbours, subculture, habit, weather, bicycle ownership, car ownership, environmental supportiveness, and congestion affect the decision to trave. The developed model (MOTIVATE) is a configurable agent-based model where social norms related to travel behaviour are used to provide a more realistic representation of the socio-ecological systems in which active commuting interventions may be deployed. The utility of this model is demonstrated using car-free days as a hypothetical intervention. In the control scenario, the odds of active travel were plausible at 0.091 (89% HPDI: [0.091, 0.091]). Compared to the control scenario, the odds of active travel were increased by 70.3% (89% HPDI: [70.3%, 70.3%]), in the intervention scenario, on non-car-free days; the effect is sustained to non-car-free days. The model is a useful tool for investigating the effect of how social networks and social norms influence the effectiveness of various interventions. If configured using real-world built environment data, it may be useful for investigating how social norms interact with the built environment to cause the emergence of commuting conventions.

Conic optimization has recently emerged as a powerful tool for designing tractable and guaranteed algorithms for non-convex polynomial optimization problems. On the one hand, tractability is crucial for efficiently solving large-scale problems and, on the other hand, strong bounds are needed to ensure high quality solutions. In this research, we investigate the strengthening of RLT relaxations of polynomial optimization problems through the addition of nine different types of constraints that are based on linear, second-order cone, and semidefinite programming to solve to optimality the instances of well established test sets of polynomial optimization problems. We describe how to design these conic constraints and their performance with respect to each other and with respect to the standard RLT relaxations. Our first finding is that the different variants of nonlinear constraints (second-order cone and semidefinite) are the best performing ones in around $50\%$ of the instances. Additionally, we present a machine learning approach to decide on the most suitable constraints to add for a given instance. The computational results show that the machine learning approach significantly outperforms each and every one of the nine individual approaches.