Fast Model Learning for the Detection of Malicious Digital Documents

Microscopic analysis of breast tissues is necessary for a definitive diagnosis of breast cancer which is the most common cancer among women. Pathology examination requires time consuming scanning through tissue images under different magnification levels to find clinical assessment clues to produce correct diagnoses. Advances in digital imaging techniques offers assessment of pathology images using computer vision and machine learning methods which could automate some of the tasks in the diagnostic pathology workflow. Such automation could be beneficial to obtain fast and precise quantification, reduce observer variability, and increase objectivity. In this work, we propose to classify breast cancer histopathology images independent of their magnifications using convolutional neural networks (CNNs). We propose two different architectures; single task CNN is used to predict malignancy and multi-task CNN is used to predict both malignancy and image magnification level simultaneously. Evaluations and comparisons with previous results are carried out on BreaKHis dataset. Experimental results show that our magnification independent CNN approach improved the performance of magnification specific model. Our results in this limited set of training data are comparable with previous state-of-the-art results obtained by hand-crafted features. However, unlike previous methods, our approach has potential to directly benefit from additional training data, and such additional data could be captured with same or different magnification levels than previous data.

In the field of confidentiality management, secret point recognition serves as an accurate and highly efficient approach. Due to the strict confidentiality requirements imposed on training data and the limited scope of data access, secret point recognition based on machine learning faces challenges in achieving good recognition results because of insufficient training data. This paper presents a secret point recognition algorithm using test-time augmentation (TTA) based on large language models. During training, a small amount of data was used to finetune the pretrained model. During inference, a large language model is used to paraphrase the input text, generating multiple expressions with different styles but the same meaning. These expressions are used as augmented data and are classified using a fine-tuned pre-trained classification model. The classification results were aggregated through an aggregation strategy to obtain the final prediction. To verify the effectiveness of the proposed method, we conducted experiments using simulated data. In the experiments, the ERNIE model was fine-tuned with a small amount of data. On this basis, TTA based on large language model was applied, and compared it with the BiLSTM and TextCNN methods. The experimental results demonstrate that this method can effectively improve the accuracy by 4.2 percent.

Fault diagnosis of the hydraulic valve using a novel semi-supervised learning method based on multi-sensor information fusion

Brain-computer interaction (BCI) and physiological computing are terms that refer to using processed neural or physiological signals to influence human interaction with computers, environment, and each other. A major challenge in developing these systems arises from the large individual differences typically seen in the neural/physiological responses. As a result, many researchers use individually-trained recognition algorithms to process this data. In order to minimize time, cost, and barriers to use, there is a need to minimize the amount of individual training data required, or equivalently, to increase the recognition accuracy without increasing the number of user-specific training samples. One promising method for achieving this is collaborative filtering, which combines training data from the individual subject with additional training data from other, similar subjects. This paper describes a successful application of a collaborative filtering approach intended for a BCI system. This approach is based on transfer learning (TL), active class selection (ACS), and a mean squared difference user-similarity heuristic. The resulting BCI system uses neural and physiological signals for automatic task difficulty recognition. TL improves the learning performance by combining a small number of user-specific training samples with a large number of auxiliary training samples from other similar subjects. ACS optimally selects the classes to generate user-specific training samples. Experimental results on 18 subjects, using both nearest neighbors and support vector machine classifiers, demonstrate that the proposed approach can significantly reduce the number of user-specific training data samples. This collaborative filtering approach will also be generalizable to handling individual differences in many other applications that involve human neural or physiological data, such as affective computing.

Entity Resolution (ER) is the problem of semi-automatically determining when two entities refer to the same underlying entity, with applications ranging from healthcare to e-commerce. Traditional ER solutions required considerable manual expertise, including domain-specific feature engineering, as well as identification and curation of training data. Recently released large language models (LLMs) provide an opportunity to make ER more seamless and domain-independent. Because of LLMs’ pre-trained knowledge, the matching step in ER can be made easier by just prompting. However, it is also well known that LLMs can pose risks, that the quality of their outputs can depend on how prompts are engineered, and that the cost of using LLMs can be significant. Unfortunately, a systematic experimental study on the effects of different prompting methods and their respective cost for solving domain-specific entity matching using LLMs, like ChatGPT, has been lacking thus far. This paper aims to address this gap by conducting such a study. We consider some relatively simple and cost-efficient ER prompt engineering methods and apply them to perform product matching on two real-world datasets widely used in the community. We select two well-known e-commerce datasets and provide extensive experimental results to show that an LLM like GPT3.5 is viable for high-performing product matching and, interestingly, that more complicated and detailed (and hence, expensive) prompting methods do not necessarily outperform simpler approaches. We provide brief discussions on qualitative and error analysis, including a study of the inter-consistency of different prompting methods to determine whether they yield stable outputs. Finally, we consider some limitations of LLMs when used as a product matcher in potential real-world e-commerce applications.

Data-driven gait prediction can provide a reference trajectory for a wide variety of simple and complex movements captured in the training data. Coordinated Movement (CM) is a data-driven approach that maps movements of the body to movements of target joints, such as the ankle and knee. We have previously shown that the performance of CM for complex activities can be improved by adding more training data. In this paper we demonstrate that performance can also be improved by 1) including a history of the target joint angles as inputs to the model and 2) dynamic reallocation of the importance of the inputs over time using a neural network technique called Attention. These modifications are applicable when additional training data is limited. We also observe that Attention can follow important events in gait over time, adding interpretability to the system.

We introduce three new Empirical Seawater Property Estimation Routines (ESPERs) capable of predicting seawater phosphate, nitrate, silicate, oxygen, total titration seawater alkalinity, total hydrogen scale pH (pHT), and total dissolved inorganic carbon (DIC) from up to 16 combinations of seawater property measurements. The routines generate estimates from neural networks (ESPER_NN), locally interpolated regressions (ESPER_LIR), or both (ESPER_Mixed). They require a salinity value and coordinate information, and benefit from additional seawater measurements if available. These routines are intended for seawater property measurement quality control and quality assessment, generating estimates for calculations that require approximate values, original science, and producing biogeochemical property context from a data set. Relative to earlier LIR routines, the updates expand their functionality, including new estimated properties and combinations of predictors, a larger training data product including new cruises from the 2020 Global Data Analysis Project data product release, and the implementation of a first‐principles approach for quantifying the impacts of anthropogenic carbon on DIC and pHT. We show that the new routines perform at least as well as existing routines, and, in some cases, outperform existing approaches, even when limited to the same training data. Given that additional training data has been incorporated into these updated routines, these updates should be considered an improvement over earlier versions. The routines are intended for all ocean depths for the interval from 1980 to ~2030 c.e., and we caution against using the routines to directly quantify surface ocean seasonality or make more distant predictions of DIC or pHT.

We consider a serious, previously-unexplored challenge facing almost all approaches to scaling up entity resolution (ER) to multiple data sources: the prohibitive cost of labeling training data for supervised learning of similarity scores for each pair of sources. While there exists a rich literature describing almost all aspects of pairwise ER, this new challenge is arising now due to the unprecedented ability to acquire and store data from online sources, interest in features driven by ER such as enriched search verticals, and the uniqueness of noisy and missing data characteristics for each source. We show on real-world and synthetic data that for state-of-the-art techniques, the reality of heterogeneous sources means that the number of labeled training data must scale quadratically in the number of sources, just to maintain constant precision/recall. We address this challenge with a brand new transfer learning algorithm which requires far less training data (or equivalently, achieves superior accuracy with the same data) and is trained using fast convex optimization. The intuition behind our approach is to adaptively share structure learned about one scoring problem with all other scoring problems sharing a data source in common. We demonstrate that our theoretically-motivated approach improves upon existing techniques for multi-source ER.

Boosting implicit discourse relation recognition with connective-based word embeddings

We show an application of artificial neural networks for local weather prediction. By employment of appropriate network structure and proper selection of input/output signals, solid results can be achieved. Our system was implemented in the local district heating company, where it was used to predict daily temperature profile with period of 15 minutes. Further, weekly and yearly profiles were predicted, and also heat consumption profiles. Whole prediction system consists of several chained neural networks and data processing modules. Training data for neural networks were collected from meteorological stations around the Košice city. Additional training data were collected by web-robots from internet from several weather forecast agencies.KeywordsPrediction SystemPrediction ModuleDistrict HeatingHeat ConsumptionSteam PowerThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Abstract. Knowledge of cloud phase (liquid, ice, mixed, etc.) is necessary to describe the radiative impact of clouds and their lifetimes, but is a property that is difficult to simulate correctly in climate models. One step towards improving those simulations is to make observations of cloud phase with sufficient accuracy to help constrain model representations of cloud processes. In this study, we outline a methodology using a basic Bayesian classifier to estimate the probabilities of cloud-phase class from Atmospheric Radiation Measurement (ARM) vertically pointing active remote sensors. The advantage of this method over previous ones is that it provides uncertainty information on the phase classification. We also test the value of including higher moments of the cloud radar Doppler spectrum than are traditionally used operationally. Using training data of known phase from the Mixed-Phase Arctic Cloud Experiment (M-PACE) field campaign, we demonstrate a proof of concept for how the method can be used to train an algorithm that identifies ice, liquid, mixed phase, and snow. Over 95 % of data are identified correctly for pure ice and liquid cases used in this study. Mixed-phase and snow cases are more problematic to identify correctly. When lidar data are not available, including additional information from the Doppler spectrum provides substantial improvement to the algorithm. This is a first step towards an operational algorithm and can be expanded to include additional categories such as drizzle with additional training data.

Due to recent technological advances in capturing and processing devices, hyperspectral imaging is becoming available for many commercial and military applications such as remote sensing, surveillance, and forest fire detection. Hyperspectral cameras provide rich information, as they capture each pixel along many frequency bands in the spectrum. The large volume of hyperspectral images as well as their high dimensionality make transmitting them over limited-bandwidth channels a challenge. To address this challenge, we present a method to prioritize the transmission of various components of hyperspectral data based on the application needs, the level of details required, and available bandwidth. This is unlike current works that mostly assume offline processing and the availability of all data beforehand. Our method jointly and optimally selects the spectral bands and their qualities to maximize the utility of the transmitted data. It also enables progressive transmission of hyperspectral data, in which approximate results are obtained with small amount of data and can be refined with additional data. This is a desirable feature for large-scale hyperspectral imaging applications. We have implemented the proposed method and compared it against the state-of-the-art in the literature using hyperspectral imaging datasets. Our experimental results show that the proposed method achieves high accuracy, transmits a small fraction of the hyperspectral data, and significantly outperforms the state-of-the-art; up to 35% improvements in accuracy was achieved.

The amount of data with which a historian works is sometimes limited by the availability of sources. When qualitative distinctions are to be compared in the context of only a small amount of data, ACTUS, a computer application originally described in this journal in 1989 by Carl B. Estabrook and George F. Estabrook, may enable a scholar to make a statistically supported interpretation in cases where he or she might otherwise have avoided interpretation, have made a purely verbal argument, or have even been misled. The present work presents ACTUS2, which implements several new hypotheses of independence.

The 2019 FEIII CALI data challenge aims at linking different representations of the same real-world entities across multiple public datasets that collect identification and activity data about small to medium enterprises (SMEs) in California. We formalize this challenge as a learning-based entity resolution (ER) task, the goal of which is to learn a high-precision and high-recall pair-wise ER model that classifies small business entity pairs into matches and non-matches. Realistic ER tasks usually involve a pipeline of laborintensive and error-prone tasks, such as data preprocesing, gathering of training data, feature engineering, and model tuning. In this task, we apply an advanced human-in-the-loop system, named SystemER, to learn ER algorithms for SME entities. Powered by active learning and via a carefully designed user interface, SystemER can learn high-quality explainable ER algorithms with low human effort, while achieving high-accuracy on the datasets provided by the FEIII CALI data challenge.

Transfer Learning and Selective data training are two of the many approaches being extensively investigated to improve the quality of Neural Machine Translation systems. This paper presents a series of experiments by applying transfer learning and selective data training for participation in the Bio-medical shared task of WMT19. We have used Information Retrieval to selectively choose related sentences from out-of-domain data and used them as additional training data using transfer learning. We also report the effect of tokenization on translation model performance.