Towards Effective Patient Simulators

In this paper we give an overview of the field of patient simulators and provide qualitative and quantitative comparison of different modeling and simulation approaches. Simulators can be used to train human caregivers but also to develop and optimize algorithms for clinical decision support applications and test and validate interventions. In this paper we introduce three novel patient simulators with different levels of representational accuracy: HeartPole, a simplistic transparent rule-based system, GraphSim, a graph-based model trained on intensive care data, and Auto-ALS—an adjusted version of an educational software package used for training junior healthcare professionals. We provide a qualitative and quantitative comparison of the previously existing as well as proposed simulators.

The Novel in the Spanish Silver Age

What distinguishes an adventure novel from a historical novel? Can the same text belong to several genres? More to one than to another? Have some existing genres been overlooked? To answer these and similar questions, José Calvo Tello combines methods from Linguistics (lexicography), Literary Studies (genre theory), and Computer Science (machine learning, natural language processing). Located in the interdisciplinary field of Digital Humanities, this study analyzes a newly developed corpus of 358 Spanish novels of the silver age (1880-1939), which includes authors like Baroja, Pardo Bazán, or Valle-Inclán. Calvo Tello's key result is a graph-based model of literary genre that reconciles recent theoretical approaches.

Uncertainty-Aware Few-Shot Image Classification

Few-shot image classification learns to recognize new categories from limited labelled data. Metric learning based approaches have been widely investigated, where a query sample is classified by finding the nearest prototype from the support set based on their feature similarities. A neural network has different uncertainties on its calculated similarities of different pairs. Understanding and modeling the uncertainty on the similarity could promote the exploitation of limited samples in few-shot optimization. In this work, we propose Uncertainty-Aware Few-Shot framework for image classification by modeling uncertainty of the similarities of query-support pairs and performing uncertainty-aware optimization. Particularly, we exploit such uncertainty by converting observed similarities to probabilistic representations and incorporate them to the loss for more effective optimization. In order to jointly consider the similarities between a query and the prototypes in a support set, a graph-based model is utilized to estimate the uncertainty of the pairs. Extensive experiments show our proposed method brings significant improvements on top of a strong baseline and achieves the state-of-the-art performance.

TEC: A Time Evolving Contextual Graph Model for Speaker State Analysis in Political Debates

Political discourses provide a forum for representatives to express their opinions and contribute towards policy making. Analyzing these discussions is crucial for recognizing possible delegates and making better voting choices in an independent nation. A politician's vote on a proposition is usually associated with their past discourses and impacted by cohesion forces in political parties. We focus on predicting a speaker's vote on a bill by augmenting linguistic models with temporal and cohesion contexts. We propose TEC, a time evolving graph based model that jointly employs links between motions, speakers, and temporal politician states. TEC outperforms competitive models, illustrating the benefit of temporal and contextual signals for predicting a politician's stance.

Multi-peak Graph-based Multi-instance Learning for Weakly Supervised Object Detection

Weakly supervised object detection (WSOD), aiming to detect objects with only image-level annotations, has become one of the research hotspots over the past few years. Recently, much effort has been devoted to WSOD for the simple yet effective architecture and remarkable improvements have been achieved. Existing approaches using multiple-instance learning usually pay more attention to the proposals individually, ignoring relation information between proposals. Besides, to obtain pseudo-ground-truth boxes for WSOD, MIL-based methods tend to select the region with the highest confidence score and regard those with small overlap as background category, which leads to mislabeled instances. As a result, these methods suffer from mislabeling instances and lacking relations between proposals, degrading the performance of WSOD. To tackle these issues, this article introduces a multi-peak graph-based model for WSOD. Specifically, we use the instance graph to model the relations between proposals, which reinforces multiple-instance learning process. In addition, a multi-peak discovery strategy is designed to avert mislabeling instances. The proposed model is trained by stochastic gradients decent optimizer using back-propagation in an end-to-end manner. Extensive quantitative and qualitative evaluations on two publicly challenging benchmarks, PASCAL VOC 2007 and PASCAL VOC 2012, demonstrate the superiority and effectiveness of the proposed approach.

Robust Discriminative Analysis Framework for Gaze and Headpose Estimation

Head movements, combined with gaze, play a fundamental role in predicting a person’s action and intention. In non-constrained head movement settings, the process is complex, and performance can degrade significantly in the presence of variation in head-pose, gaze position, occlusion and ambient illumination. In this thesis, a framework is therefore proposed to fuse and combine head-pose and gaze information to obtain more robust and accurate gaze estimation. Specific contributions include: the development of a newly developed graph-based model for pupil localization and accurate estimation of the pupil center; the proposal of a novel iris region descriptor feature using quadtree decomposition, that works together with pupil localization for gaze estimation; the proposal of kernel-based extensions and enhancements to a fusion mechanism known as Discriminative Multiple Canonical Correlation Analysis (DMCCA) for fusing features (proposed and traditional) together, to generate a refined, high quality feature set for classification; and the newly developed methodology of head-pose features based on quadtree decompositions and geometrical moments, to better integrate roll, yaw, pitch and jawline into the overall estimation framework. The experimental results of the proposed framework demonstrate robustness against variations in illumination, occlusion, head-pose and is calibration free. The proposed framework was validated on several datasets and scored: 4.5° using MPII, 4.4° using Cave, 4.8° using EYEDIAP, 5.0° using ACS, 4.1° using OSLO and 4.5° using UULM datasets respectively.

Robust Discriminative Analysis Framework for Gaze and Headpose Estimation

Head movements, combined with gaze, play a fundamental role in predicting a person’s action and intention. In non-constrained head movement settings, the process is complex, and performance can degrade significantly in the presence of variation in head-pose, gaze position, occlusion and ambient illumination. In this thesis, a framework is therefore proposed to fuse and combine head-pose and gaze information to obtain more robust and accurate gaze estimation. Specific contributions include: the development of a newly developed graph-based model for pupil localization and accurate estimation of the pupil center; the proposal of a novel iris region descriptor feature using quadtree decomposition, that works together with pupil localization for gaze estimation; the proposal of kernel-based extensions and enhancements to a fusion mechanism known as Discriminative Multiple Canonical Correlation Analysis (DMCCA) for fusing features (proposed and traditional) together, to generate a refined, high quality feature set for classification; and the newly developed methodology of head-pose features based on quadtree decompositions and geometrical moments, to better integrate roll, yaw, pitch and jawline into the overall estimation framework. The experimental results of the proposed framework demonstrate robustness against variations in illumination, occlusion, head-pose and is calibration free. The proposed framework was validated on several datasets and scored: 4.5° using MPII, 4.4° using Cave, 4.8° using EYEDIAP, 5.0° using ACS, 4.1° using OSLO and 4.5° using UULM datasets respectively.

Movie recommender systems using hybrid model based on graphs with co-rated, genre, and closed caption features

A movie recommendation is a long-standing challenge. Figuring out the viewer’s interest in movies is still a problem since a huge number of movies are released in no time. In the meantime, people cannot enjoy all available new releases or unseen movies due to their limited time. They also still need to choose which movies to watch when they have spare time. This situation is not good for the movie business too. In order to satisfy people in choosing what movies to watch and to boost movie sales, a system that can recommend suitable movies is required, either unseen in the past or new releases. This paper focuses on the hybrid approach, a combination of content-based and collaborative filtering, using a graph-based model. This hybrid approach is proposed to overcome the drawbacks of combination in the content-based and collaborative filtering. The graph database, Neo4j is used to store the collaborative features, such as movies with its genres, and ratings. Since the movie’s closed caption is rarely considered to be used in a recommendation, the proposed method evaluates the impact of using this syntactic feature. From the early test, the combination of collaborative filtering and content-based using closed caption gives a slightly better result than without closed caption, especially in finding similar movies such as sequel or prequel.