Contextual bandits algorithms have become essential in real-world user interaction problems in recent years. However, these algorithms rely on context as attribute value representation, which makes them unfeasible for real-world domains like social networks are inherently relational. We propose Relational Boosted Bandits(RB2), acontextual bandits algorithm for relational domains based on (relational) boosted trees. RB2 enables us to learn interpretable and explainable models due to the more descriptive nature of the relational representation. We empirically demonstrate the effectiveness and interpretability of RB2 on tasks such as link prediction, relational classification, and recommendations.
The popularity of e-commerce platforms continues to grow. Being able to understand, and predict customer behavior is essential for customizing the user experience through personalized result presentations, recommendations, and special offers. Previous work has considered a broad range of prediction models as well as features inferred from clickstream data to record session characteristics, and features inferred from user data to record customer characteristics. So far, most previous work in the area of purchase prediction has focused on known customers, largely ignoring anonymous sessions, i.e., sessions initiated by a non-logged-in or unrecognized customer. However, in the de-identified data from a large European e-commerce platform available to us, more than 50% of the sessions start as anonymous sessions. In this paper, we focus on purchase prediction for both anonymous and identified sessions on an e-commerce platform. We start with a descriptive analysis of purchase vs. non-purchase sessions. This analysis informs the definition of a feature-based model for purchase prediction for anonymous sessions and identified sessions; our models consider a range of session-based features for anonymous sessions, such as the channel type, the number of visited pages, and the device type. For identified user sessions, our analysis points to customer history data as a valuable discriminator between purchase and non-purchase sessions. Based on our analysis, we build two types of predictors: (1) a predictor for anonymous that beats a production-ready predictor by over 17.54% F1; and (2) a predictor for identified customers that uses session data as well as customer history and achieves an F1 of 96.20%. Finally, we discuss the broader practical implications of our findings.
Probabilistic solvers for ordinary differential equations assign a posterior measure to the solution of an initial value problem. The joint covariance of this distribution provides an estimate of the (global) approximation error. The contraction rate of this error estimate as a function of the solver's step size identifies it as a well-calibrated worst-case error, but its explicit numerical value for a certain step size is not automatically a good estimate of the explicit error. Addressing this issue, we introduce, discuss, and assess several probabilistically motivated ways to calibrate the uncertainty estimate. Numerical experiments demonstrate that these calibration methods interact efficiently with adaptive step-size selection, resulting in descriptive, and efficiently computable posteriors. We demonstrate the efficiency of the methodology by benchmarking against the classic, widely used Dormand-Prince 4/5 Runge-Kutta method.
Automatic detecting anomalous regions in images of objects or textures without priors of the anomalies is challenging, especially when the anomalies appear in very small areas of the images, making difficult-to-detect visual variations, such as defects on manufacturing products. This paper proposes an effective unsupervised anomaly segmentation approach that can detect and segment out the anomalies in small and confined regions of images. Concretely, we develop a multi-scale regional feature generator that can generate multiple spatial context-aware representations from pre-trained deep convolutional networks for every subregion of an image. The regional representations not only describe the local characteristics of corresponding regions but also encode their multiple spatial context information, making them discriminative and very beneficial for anomaly detection. Leveraging these descriptive regional features, we then design a deep yet efficient convolutional autoencoder and detect anomalous regions within images via fast feature reconstruction. Our method is simple yet effective and efficient. It advances the state-of-the-art performances on several benchmark datasets and shows great potential for real applications.
Cloud-based Deep Neural Network (DNN) applications that make latency-sensitive inference are becoming an indispensable part of Industry 4.0. Due to the multi-tenancy and resource heterogeneity, both inherent to the cloud computing environments, the inference time of DNN-based applications are stochastic. Such stochasticity, if not captured, can potentially lead to low Quality of Service (QoS) or even a disaster in critical sectors, such as Oil and Gas industry. To make Industry 4.0 robust, solution architects and researchers need to understand the behavior of DNN-based applications and capture the stochasticity exists in their inference times. Accordingly, in this study, we provide a descriptive analysis of the inference time from two perspectives. First, we perform an application-centric analysis and statistically model the execution time of four categorically different DNN applications on both Amazon and Chameleon clouds. Second, we take a resource-centric approach and analyze a rate-based metric in form of Million Instruction Per Second (MIPS) for heterogeneous machines in the cloud. This non-parametric modeling, achieved via Jackknife and Bootstrap re-sampling methods, provides the confidence interval of MIPS for heterogeneous cloud machines. The findings of this research can be helpful for researchers and cloud solution architects to develop solutions that are robust against the stochastic nature of the inference time of DNN applications in the cloud and can offer a higher QoS to their users and avoid unintended outcomes.
Current summarization systems yield generic summaries that are disconnected from users' preferences and expectations. To address this limitation, we present CTRLsum, a novel framework for controllable summarization. Our approach enables users to control multiple aspects of generated summaries by interacting with the summarization system through textual input in the form of a set of keywords or descriptive prompts. Using a single unified model, CTRLsum is able to achieve a broad scope of summary manipulation at inference time without requiring additional human annotations or pre-defining a set of control aspects during training. We quantitatively demonstrate the effectiveness of our approach on three domains of summarization datasets and five control aspects: 1) entity-centric and 2) length-controllable summarization, 3) contribution summarization on scientific papers, 4) invention purpose summarization on patent filings, and 5) question-guided summarization on news articles in a reading comprehension setting. Moreover, when used in a standard, uncontrolled summarization setting, CTRLsum achieves state-of-the-art results on the CNN/DailyMail dataset. Code and model checkpoints are available at https://github.com/salesforce/ctrl-sum
Humans are able to perceive, understand and reason about causal events. Developing models with similar physical and causal understanding capabilities is a long-standing goal of artificial intelligence. As a step towards this direction, we introduce CRAFT, a new video question answering dataset that requires causal reasoning about physical forces and object interactions. It contains 58K video and question pairs that are generated from 10K videos from 20 different virtual environments, containing various objects in motion that interact with each other and the scene. Two question categories in CRAFT include previously studied descriptive and counterfactual questions. Additionally, inspired by the Force Dynamics Theory in cognitive linguistics, we introduce a new causal question category that involves understanding the causal interactions between objects through notions like cause, enable, and prevent. Our results show that even though the questions in CRAFT are easy for humans, the tested baseline models, including existing state-of-the-art methods, do not yet deal with the challenges posed in our benchmark.
In the last years, AI safety gained international recognition in the light of heterogeneous safety-critical and ethical issues that risk overshadowing the broad beneficial impacts of AI. In this context, the implementation of AI observatory endeavors represents one key research direction. This paper motivates the need for an inherently transdisciplinary AI observatory approach integrating diverse retrospective and counterfactual views. We delineate aims and limitations while providing hands-on-advice utilizing concrete practical examples. Distinguishing between unintentionally and intentionally triggered AI risks with diverse socio-psycho-technological impacts, we exemplify a retrospective descriptive analysis followed by a retrospective counterfactual risk analysis. Building on these AI observatory tools, we present near-term transdisciplinary guidelines for AI safety. As further contribution, we discuss differentiated and tailored long-term directions through the lens of two disparate modern AI safety paradigms. For simplicity, we refer to these two different paradigms with the terms artificial stupidity (AS) and eternal creativity (EC) respectively. While both AS and EC acknowledge the need for a hybrid cognitive-affective approach to AI safety and overlap with regard to many short-term considerations, they differ fundamentally in the nature of multiple envisaged long-term solution patterns. By compiling relevant underlying contradistinctions, we aim to provide future-oriented incentives for constructive dialectics in practical and theoretical AI safety research.
Image captioning models generally lack the capability to take into account user interest, and usually default to global descriptions that try to balance readability, informativeness, and information overload. On the other hand, VQA models generally lack the ability to provide long descriptive answers, while expecting the textual question to be quite precise. We present a method to control the concepts that an image caption should focus on, using an additional input called the guiding text that refers to either groundable or ungroundable concepts in the image. Our model consists of a Transformer-based multimodal encoder that uses the guiding text together with global and object-level image features to derive early-fusion representations used to generate the guided caption. While models trained on Visual Genome data have an in-domain advantage of fitting well when guided with automatic object labels, we find that guided captioning models trained on Conceptual Captions generalize better on out-of-domain images and guiding texts. Our human-evaluation results indicate that attempting in-the-wild guided image captioning requires access to large, unrestricted-domain training datasets, and that increased style diversity (even without increasing the number of unique tokens) is a key factor for improved performance.
We introduce the task of dense captioning in 3D scans from commodity RGB-D sensors. As input, we assume a point cloud of a 3D scene; the expected output is the bounding boxes along with the descriptions for the underlying objects. To address the 3D object detection and description problems, we propose Scan2Cap, an end-to-end trained method, to detect objects in the input scene and describe them in natural language. We use an attention mechanism that generates descriptive tokens while referring to the related components in the local context. To reflect object relations (i.e. relative spatial relations) in the generated captions, we use a message passing graph module to facilitate learning object relation features. Our method can effectively localize and describe 3D objects in scenes from the ScanRefer dataset, outperforming 2D baseline methods by a significant margin (27.61% [email omitted]).
Human beings possess the most sophisticated computational machinery in the known universe. We can understand language of rich descriptive power, and communicate in the same environment with astonishing clarity. Two of the many contributors to the interest in natural language - the properties of Compositionality and Structure Dependence, are well documented, and offer a vast space to ask interesting modelling questions. The first step to begin answering these questions is to ground verbal theory in formal terms. Drawing on linguistics and set theory, a formalisation of these ideas is presented in the first half of this thesis. We see how cognitive systems that process language need to have certain functional constraints, viz. time based, incremental operations that rely on a structurally defined domain. The observations that result from analysing this formal setup are examined as part of a modelling exercise. Using the advances of word embedding techniques, a model of relational learning is simulated with a custom dataset to demonstrate how a time based role-filler binding mechanism satisfies some of the constraints described in the first section. The model's ability to map structure, along with its symbolic-connectionist architecture makes for a cognitively plausible implementation. The formalisation and simulation are together an attempt to recognise the constraints imposed by linguistic theory, and explore the opportunities presented by a cognitive model of relation learning to realise these constraints.
Microfinance provides strength to boost the economic activities of low-income earners and thus contributes to eradication of poverty. However, microfinance institutions face stringent competition from commercial banks; the growth of microloan activities of commercial banks may confront microfinance institutions with increased competition for borrowers. In Kenya, the micro finance sector has extremely high competition indicated by the shifting market share and profitability. This study sought to examine the determinants of financial performance of Microfinance banks in Kenya. The study adopted a descriptive research design and used secondary data from 7 Microfinance banks for a period of 5 years from 2011 to 2015. The data collected was analyzed using correlation and regression analysis. The study found a positive and statistically significant relationship between operational efficiency, capital adequacy, firm size and financial performance of microfinance banks in Kenya. However, the study found an insignificant negative relationship between liquidity risk, credit risk and financial performance of microfinance banks in Kenya. The study concluded that there is direct relationship between operational efficiency, capital adequacy, firm size and financial performance of microfinance banks in Kenya.
Small and Medium Enterprises (SMEs) access to external finance is an issue of significant research interest to academicians. Commercial banks consider many SMEs not to be credit worthy because of their inability to meet some banking requirements. Hence, the objective of this study was to investigate what determines lending to SMEs by commercial banks in Kenya. To achieve the study objectives, a descriptive research design was employed. The study undertook a census of the 43 commercial banks in Kenya, with full data being obtained for 36 institutions. The study used secondary data from the annual published reports of commercial banks in Kenya for a period of 5 years from 2010-2014. The data collected was analyzed through the multiple linear regression using the Statistical Package for Social Studies version 20.The study established that bank size and liquidity significantly influences (positively and negatively, respectively) lending to SMEs by commercial banks in Kenya while credit risk and interest rates have no significant influence on lending to SMEs by commercial banks in Kenya. The study recommends that lending to SMEs by commercial banks in Kenya be enhanced by adopting policies that grow the commercial banks.
The radar equation is the fundamental mathematical modelling of the basic function of a radar system. Moreover there are many versions of the radar equation that correspond to particular radar operations, like a low PRF, a high PRF or a surveillance mode. In many cases all these expressions of the radar equation exist in their combined forms giving little information to the actual physics and signal geometry between the radar and target involved in the process. In this case study we divide the radar equation into its major steps and present a descriptive mathematical modelling of the radar and other related equations utilizing the free space loss and the target gain concepts for simulating the effect of a white noise jammer to an adversary radar. We believe that this work will be particularly beneficial to instructors of radar courses and to radar simulation engineers because of its analytical block approach of the main equations related to radar and electronic warfare fields. Finally this work falls under the field of predictive dynamics for radar systems using mathematical modelling techniques.
This project aims at creating an investment device to help investors determine which real estate units have a higher return to investment in Madrid. To do so, we gather data from Idealista.com, a real estate web-page with millions of real estate units across Spain, Italy and Portugal. In this preliminary version, we present the road map on how we gather the data; descriptive statistics of the 8,121 real estate units gathered (rental and sale); build a return index based on the difference in prices of rental and sale units(per neighbourhood and size) and introduce machine learning algorithms for rental real estate price prediction.
Background: Solitary fibrous tumor is the second primary malignancy of the pleura and can reach up to 39cm in diameter; to be referred to as giant it must occupy at least 40% of the affected hemithorax. Although this tumor usually shows a benign behavior malignancy criteria have been described. The aim of the study was to assess the initial evaluation diagnostic procedures surgical management treatment outcome and prognosis. Methods: We performed a descriptive observational longitudinal and retrospective study from 2002 to 2006 on patients who underwent surgery with a diagnosis of giant solitary fibrous tumor of the pleura. Results: Six patients were included; 83.3% were females. Mean age was 48 years. All patients were symptomatic mainly dyspnea cough and chest pain; 66.7% were left-sided. Preoperative angiography and embolization were performed in 83.3% cases with successful surgical resection. The predominant blood supply was derived from the internal mammalian artery. Intraoperative complication rate was 17%. A vascular pedicle was found in 66.7%. The largest lesion was 40cm in diameter and weighed 4500g. Only one case showed high mitotic activity. Mean follow-up to date is 14 months. Conclusions: Symptomatology found was consistent with previous reports but in higher percentages. Accurate diagnosis is critical because surgical resection involves a potential cure; long-term follow-up is mandatory. Preoperative embolization is recommended due to tumor size.
Inertia and context-dependent choice effects are well-studied classes of behavioural phenomena. While much is known about these effects in isolation, little is known about whether one of them "dominates" the other when both can potentially be present. Knowledge of any such dominance is relevant for effective choice architecture and descriptive modelling. We initiate this empirical investigation with a between-subjects lab experiment in which each subject made a single decision over two or three money lotteries. Our experiment was designed to test for dominance between *status quo bias* and the *decoy effect*. We find strong evidence for status quo bias and no evidence for the decoy effect. We also find that status quo bias can be powerful enough so that, at the aggregate level, a fraction of subjects switch from being risk-averse to being risk-seeking. Survey evidence suggests that this is due to subjects focusing on the maximum possible amount when the risky lottery is the default and on the highest probability of winning the biggest possible reward when there is no default. The observed reversal in risk attitudes is explainable by a large class of Koszegi-Rabin (2006) reference-dependent preferences.
Time-frequency distributions (TFDs) play a vital role in providing descriptive analysis of non-stationary signals involved in realistic scenarios. It is well known that low time-frequency (TF) resolution and the emergency of cross-terms (CTs) are two main issues, which make it difficult to analyze and interpret practical signals using TFDs. In order to address these issues, we propose the U-Net aided iterative shrinkage-thresholding algorithm (U-ISTA) for reconstructing a near-ideal TFD by exploiting structured sparsity in signal TF domain. Specifically, the signal ambiguity function is firstly compressed, followed by unfolding the ISTA as a recurrent neural network. To consider continuously distributed characteristics of signals, a structured sparsity constraint is incorporated into the unfolded ISTA by regarding the U-Net as an adaptive threshold block, in which structure-aware thresholds are learned from enormous training data to exploit the underlying dependencies among neighboring TF coefficients. The proposed U-ISTA model is trained by both non-overlapped and overlapped synthetic signals including closely and far located non-stationary components. Experimental results demonstrate that the robust U-ISTA achieves superior performance compared with state-of-the-art algorithms, and gains a high TF resolution with CTs greatly eliminated even in low signal-to-noise ratio (SNR) environments.
The article reviews the history of well-being to gauge how subjective question surveys can improve our understanding of well-being in Mexico. The research uses data at the level of the 32 federal entities or States, taking advantage of the heterogeneity in development indicator readings between and within geographical areas, the product of socioeconomic inequality. The data come principally from two innovative subjective questionnaires, BIARE and ENVIPE, which intersect in their fully representative state-wide applications in 2014, but also from conventional objective indicator sources such as the HDI and conventional surveys. This study uses two approaches, a descriptive analysis of a state-by-state landscape of indicators, both subjective and objective, in an initial search for stand-out well-being patterns, and an econometric study of a large selection of mainly subjective indicators inspired by theory and the findings of previous Mexican research. Descriptive analysis confirms that subjective well-being correlates strongly with and complements objective data, providing interesting directions for analysis. The econometrics literature indicates that happiness increases with income and satisfying of material needs as theory suggests, but also that Mexicans are relatively happy considering their mediocre incomes and high levels of insecurity, the last of which, by categorizing according to satisfaction with life, can be shown to impact poorer people disproportionately. The article suggests that well-being is a complex, multidimensional construct which can be revealed by using exploratory multi-regression and partial correlations models which juxtapose subjective and objective indicators.
In the wake of the COVID-19 virus outbreak, a brief phenomenological (descriptive, comparative) analysis of the dynamics of the disease spread among different countries is presented. Results indicate that the infection spread dynamics is currently the most pronounced in the USA (confirmed cases are currently doubling every 2.16 days, with a decreasing doubling time tendency), while other countries with the most confirmed cases show different values, and tendencies. The reported number of deaths is currently doubled every 2.28 days in Germany, 2.56 days in France, 2.57 days in Switzerland, 2.59 days in France, and 2.62 days in USA, while only France and USA are currently exhibiting further acceleration (diminishing doubling time).