We present deep significance clustering (DICE), a framework for jointly performing representation learning and clustering for "outcome-aware" stratification. DICE is intended to generate cluster membership that may be used to categorize a population by individual risk level for a targeted outcome. Following the representation learning and clustering steps, we embed the objective function in DICE with a constraint which requires a statistically significant association between the outcome and cluster membership of learned representations. DICE further includes a neural architecture search step to maximize both the likelihood of representation learning and outcome classification accuracy with cluster membership as the predictor. To demonstrate its utility in medicine for patient risk-stratification, the performance of DICE was evaluated using two datasets with different outcome ratios extracted from real-world electronic health records. Outcomes are defined as acute kidney injury (30.4\%) among a cohort of COVID-19 patients, and discharge disposition (36.8\%) among a cohort of heart failure patients, respectively. Extensive results demonstrate that DICE has superior performance as measured by the difference in outcome distribution across clusters, Silhouette score, Calinski-Harabasz index, and Davies-Bouldin index for clustering, and Area under the ROC Curve (AUC) for outcome classification compared to several baseline approaches.

In this paper, we study the importance of pruning in Deep Networks (DNs) and motivate it based on the current absence of data aware weight initialization. Current DN initializations, focusing primarily at maintaining first order statistics of the feature maps through depth, force practitioners to overparametrize a model in order to reach high performances. This overparametrization can then be pruned a posteriori, leading to a phenomenon known as "winning tickets". However, the pruning literature still relies on empirical investigations, lacking a theoretical understanding of (1) how pruning affects the decision boundary, (2) how to interpret pruning, (3) how to design principled pruning techniques, and (4) how to theoretically study pruning. To tackle those questions, we propose to employ recent advances in the theoretical analysis of Continuous Piecewise Affine (CPA) DNs. From this viewpoint, we can study the DNs' input space partitioning and detect the early-bird (EB) phenomenon, guide practitioners by identifying when to stop the first training step, provide interpretability into current pruning techniques, and develop a principled pruning criteria towards efficient DN training. Finally, we conduct extensive experiments to show the effectiveness of the proposed spline pruning criteria in terms of both layerwise and global pruning over state-of-the-art pruning methods.

How do we decide the fair value of individual classifiers in an ensemble model? We introduce a new class of transferable utility cooperative games to answer this question. The players in ensemble games are pre-trained binary classifiers that collaborate in an ensemble to correctly label points from a dataset. We design Troupe a scalable algorithm that designates payoffs to individual models based on the Shapley value of those in the ensemble game. We show that the approximate Shapley value of classifiers in these games is an adequate measure for selecting a subgroup of highly predictive models. In addition, we introduce the Shapley entropy a new metric to quantify the heterogeneity of machine learning ensembles when it comes to model quality. We analytically prove that our Shapley value approximation algorithm is accurate and scales to large ensembles and big data. Experimental results on graph classification tasks establish that Troupe gives precise estimates of the Shapley value in ensemble games. We demonstrate that the Shapley value can be used for pruning large ensembles, show that complex classifiers have a prime role in correct and incorrect classification decisions, and provide evidence that adversarial models receive a low valuation.

As interest in autonomous driving increases, efforts are being made to meet requirements for the high-level automation of vehicles. In this context, the functionality inside the vehicle cabin plays a key role in ensuring a safe and pleasant journey for driver and passenger alike. At the same time, recent advances in the field of artificial intelligence (AI) have enabled a whole range of new applications and assistance systems to solve automated problems in the vehicle cabin. This paper presents a thorough survey on existing work that utilizes AI methods for use-cases inside the driving cabin, focusing, in particular, on application scenarios related to (1) driving safety and (2) driving comfort. Results from the surveyed works show that AI technology has a promising future in tackling in-cabin tasks within the autonomous driving aspect.

Objectives: Functional connectivity triggered by naturalistic stimulus (e.g., movies) and machine learning techniques provide a great insight in exploring the brain functions such as fluid intelligence. However, functional connectivity are considered to be multi-layered, while traditional machine learning based on individual models not only are limited in performance, but also fail to extract multi-dimensional and multi-layered information from brain network. Methods: In this study, inspired by multi-layer brain network structure, we propose a new method namely Weighted Ensemble-model and Network Analysis, which combines the machine learning and graph theory for improved fluid intelligence prediction. Firstly, functional connectivity analysis and graphical theory were jointly employed. The functional connectivity and graphical indices computed using the preprocessed fMRI data were then all fed into auto-encoder parallelly for feature extraction to predict the fluid intelligence. In order to improve the performance, tree regression and ridge regression model were automatically stacked and fused with weighted values. Finally, layers of auto-encoder were visualized to better illustrate the connectome patterns, followed by the evaluation of the performance to justify the mechanism of brain functions. Results: Our proposed methods achieved best performance with 3.85 mean absolute deviation, 0.66 correlation coefficient and 0.42 R-squared coefficient, outperformed other state-of-the-art methods. It is also worth noting that, the optimization of the biological pattern extraction was automated though the auto-encoder algorithm. Conclusion: The proposed method not only outperforming the state-of-the-art reports, but also able to effectively capturing the biological patterns from functional connectivity during naturalistic movies state for potential clinical explorations.

Developing human-like conversational agents is a prime area in HCI research and subsumes many tasks. Predicting listener backchannels is one such actively-researched task. While many studies have used different approaches for backchannel prediction, they all have depended on manual annotations for a large dataset. This is a bottleneck impacting the scalability of development. To this end, we propose using semi-supervised techniques to automate the process of identifying backchannels, thereby easing the annotation process. To analyze our identification module's feasibility, we compared the backchannel prediction models trained on (a) manually-annotated and (b) semi-supervised labels. Quantitative analysis revealed that the proposed semi-supervised approach could attain 95% of the former's performance. Our user-study findings revealed that almost 60% of the participants found the backchannel responses predicted by the proposed model more natural. Finally, we also analyzed the impact of personality on the type of backchannel signals and validated our findings in the user-study.

Platelet products are both expensive and have very short shelf lives. As usage rates for platelets are highly variable, the effective management of platelet demand and supply is very important yet challenging. The primary goal of this paper is to present an efficient forecasting model for platelet demand at Canadian Blood Services (CBS). To accomplish this goal, four different demand forecasting methods, ARIMA (Auto Regressive Moving Average), Prophet, lasso regression (least absolute shrinkage and selection operator) and LSTM (Long Short-Term Memory) networks are utilized and evaluated. We use a large clinical dataset for a centralized blood distribution centre for four hospitals in Hamilton, Ontario, spanning from 2010 to 2018 and consisting of daily platelet transfusions along with information such as the product specifications, the recipients' characteristics, and the recipients' laboratory test results. This study is the first to utilize different methods from statistical time series models to data-driven regression and a machine learning technique for platelet transfusion using clinical predictors and with different amounts of data. We find that the multivariate approaches have the highest accuracy in general, however, if sufficient data are available, a simpler time series approach such as ARIMA appears to be sufficient. We also comment on the approach to choose clinical indicators (inputs) for the multivariate models.

Graph-based causal discovery methods aim to capture conditional independencies consistent with the observed data and differentiate causal relationships from indirect or induced ones. Successful construction of graphical models of data depends on the assumption of causal sufficiency: that is, that all confounding variables are measured. When this assumption is not met, learned graphical structures may become arbitrarily incorrect and effects implied by such models may be wrongly attributed, carry the wrong magnitude, or mis-represent direction of correlation. Wide application of graphical models to increasingly less curated "big data" draws renewed attention to the unobserved confounder problem. We present a novel method that aims to control for the latent space when estimating a DAG by iteratively deriving proxies for the latent space from the residuals of the inferred model. Under mild assumptions, our method improves structural inference of Gaussian graphical models and enhances identifiability of the causal effect. In addition, when the model is being used to predict outcomes, it un-confounds the coefficients on the parents of the outcomes and leads to improved predictive performance when out-of-sample regime is very different from the training data. We show that any improvement of prediction of an outcome is intrinsically capped and cannot rise beyond a certain limit as compared to the confounded model. We extend our methodology beyond GGMs to ordinal variables and nonlinear cases. Our R package provides both PCA and autoencoder implementations of the methodology, suitable for GGMs with some guarantees and for better performance in general cases but without such guarantees.

Mixed membership problem for undirected network has been well studied in network analysis recent years. However, the more general case of mixed membership for directed network remains a challenge. Here, we propose an interpretable model: bipartite mixed membership stochastic blockmodel (BiMMSB for short) for directed mixed membership networks. BiMMSB allows that row nodes and column nodes of the adjacency matrix can be different and these nodes may have distinct community structure in a directed network. We also develop an efficient spectral algorithm called BiMPCA to estimate the mixed memberships for both row nodes and column nodes in a directed network. We show that the approach is asymptotically consistent under BiMMSB. We demonstrate the advantages of BiMMSB with applications to a small-scale simulation study, the directed Political blogs network and the Papers Citations network.

Bayesian Optimization is a popular tool for tuning algorithms in automatic machine learning (AutoML) systems. Current state-of-the-art methods leverage Random Forests or Gaussian processes to build a surrogate model that predicts algorithm performance given a certain set of hyperparameter settings. In this paper, we propose a new surrogate model based on gradient boosting, where we use quantile regression to provide optimistic estimates of the performance of an unobserved hyperparameter setting, and combine this with a distance metric between unobserved and observed hyperparameter settings to help regulate exploration. We demonstrate empirically that the new method is able to outperform some state-of-the art techniques across a reasonable sized set of classification problems.