Medication recommendation targets to provide a proper set of medicines according to patients' diagnoses, which is a critical task in clinics. Currently, the recommendation is manually conducted by doctors. However, for complicated cases, like patients with multiple diseases at the same time, it's difficult to propose a considerate recommendation even for experienced doctors. This urges the emergence of automatic medication recommendation which can help treat the diagnosed diseases without causing harmful drug-drug interactions.Due to the clinical value, medication recommendation has attracted growing research interests.Existing works mainly formulate medication recommendation as a multi-label classification task to predict the set of medicines. In this paper, we propose the Conditional Generation Net (COGNet) which introduces a novel copy-or-predict mechanism to generate the set of medicines. Given a patient, the proposed model first retrieves his or her historical diagnoses and medication recommendations and mines their relationship with current diagnoses. Then in predicting each medicine, the proposed model decides whether to copy a medicine from previous recommendations or to predict a new one. This process is quite similar to the decision process of human doctors. We validate the proposed model on the public MIMIC data set, and the experimental results show that the proposed model can outperform state-of-the-art approaches.
Human activity discovery aims to distinguish the activities performed by humans, without any prior information of what defines each activity. Most methods presented in human activity recognition are supervised, where there are labeled inputs to train the system. In reality, it is difficult to label data because of its huge volume and the variety of activities performed by humans. In this paper, a novel unsupervised approach is proposed to perform human activity discovery in 3D skeleton sequences. First, important frames are selected based on kinetic energy. Next, the displacement of joints, set of statistical, angles, and orientation features are extracted to represent the activities information. Since not all extracted features have useful information, the dimension of features is reduced using PCA. Most human activity discovery proposed are not fully unsupervised. They use pre-segmented videos before categorizing activities. To deal with this, we used the fragmented sliding time window method to segment the time series of activities with some overlapping. Then, activities are discovered by a novel hybrid particle swarm optimization with a Gaussian mutation algorithm to avoid getting stuck in the local optimum. Finally, k-means is applied to the outcome centroids to overcome the slow rate of PSO. Experiments on three datasets have been presented and the results show the proposed method has superior performance in discovering activities in all evaluation parameters compared to the other state-of-the-art methods and has increased accuracy of at least 4 % on average. The code is available here: https://github.com/parhamhadikhani/Human-Activity-Discovery-HPGMK
Video frame interpolation, which aims to synthesize non-exist intermediate frames in a video sequence, is an important research topic in computer vision. Existing video frame interpolation methods have achieved remarkable results under specific assumptions, such as instant or known exposure time. However, in complicated real-world situations, the temporal priors of videos, i.e. frames per second (FPS) and frame exposure time, may vary from different camera sensors. When test videos are taken under different exposure settings from training ones, the interpolated frames will suffer significant misalignment problems. In this work, we solve the video frame interpolation problem in a general situation, where input frames can be acquired under uncertain exposure (and interval) time. Unlike previous methods that can only be applied to a specific temporal prior, we derive a general curvilinear motion trajectory formula from four consecutive sharp frames or two consecutive blurry frames without temporal priors. Moreover, utilizing constraints within adjacent motion trajectories, we devise a novel optical flow refinement strategy for better interpolation results. Finally, experiments demonstrate that one well-trained model is enough for synthesizing high-quality slow-motion videos under complicated real-world situations. Codes are available on https://github.com/yjzhang96/UTI-VFI.
Mitotic count (MC) is an important histological parameter for cancer diagnosis and grading, but the manual process for obtaining MC from whole-slide histopathological images is very time-consuming and prone to error. Therefore, deep learning models have been proposed to facilitate this process. Existing approaches utilize a two-stage pipeline: the detection stage for identifying the locations of potential mitotic cells and the classification stage for refining prediction confidences. However, this pipeline formulation can lead to inconsistencies in the classification stage due to the poor prediction quality of the detection stage and the mismatches in training data distributions between the two stages. In this study, we propose a Refine Cascade Network (ReCasNet), an enhanced deep learning pipeline that mitigates the aforementioned problems with three improvements. First, window relocation was used to reduce the number of poor quality false positives generated during the detection stage. Second, object re-cropping was performed with another deep learning model to adjust poorly centered objects. Third, improved data selection strategies were introduced during the classification stage to reduce the mismatches in training data distributions. ReCasNet was evaluated on two large-scale mitotic figure recognition datasets, canine cutaneous mast cell tumor (CCMCT) and canine mammary carcinoma (CMC), which resulted in up to 4.8% percentage point improvements in the F1 scores for mitotic cell detection and 44.1% reductions in mean absolute percentage error (MAPE) for MC prediction. Techniques that underlie ReCasNet can be generalized to other two-stage object detection networks and should contribute to improving the performances of deep learning models in broad digital pathology applications.
Today, Neural Networks are the basis of breakthroughs in virtually every technical domain. Their application to accelerators has recently resulted in better performance and efficiency in these systems. At the same time, the increasing hardware failures due to the latest (shrinked) semiconductor technology needs to be addressed. Since accelerator systems are often used to back time-critical applications such as self-driving cars or medical diagnosis applications, these hardware failures must be eliminated. Our research evaluates these failures from a systemic point of view. Based on our results, we find critical results for the system reliability enhancement and we further put forth an efficient method to avoid these failures with minimal hardware overhead.
This paper investigates the massive connectivity of low Earth orbit (LEO) satellite-based Internet-of-Things (IoT) for seamless global coverage. We propose to integrate the grant-free non-orthogonal multiple access (GF-NOMA) paradigm with the emerging orthogonal time frequency space (OTFS) modulation to accommodate the massive IoT access, and mitigate the long round-trip latency and severe Doppler effect of terrestrial-satellite links (TSLs). On this basis, we put forward a two-stage successive active terminal identification (ATI) and channel estimation (CE) scheme as well as a low-complexity multi-user signal detection (SD) method. Specifically, at the first stage, the proposed training sequence aided OTFS (TS-OTFS) data frame structure facilitates the joint ATI and coarse CE, whereby both the traffic sparsity of terrestrial IoT terminals and the sparse channel impulse response are leveraged for enhanced performance. Moreover, based on the single Doppler shift property for each TSL and sparsity of delay-Doppler domain channel, we develop a parametric approach to further refine the CE performance. Finally, a least square based parallel time domain SD method is developed to detect the OTFS signals with relatively low complexity. Simulation results demonstrate the superiority of the proposed methods over the state-of-the-art solutions in terms of ATI, CE, and SD performance confronted with the long round-trip latency and severe Doppler effect.
Video-to-video synthesis is a challenging problem aiming at learning a translation function between a sequence of semantic maps and a photo-realistic video depicting the characteristics of a driving video. We propose a head-to-head system of our own implementation capable of fully transferring the human head 3D pose, facial expressions and eye gaze from a source to a target actor, while preserving the identity of the target actor. Our system produces high-fidelity, temporally-smooth and photo-realistic synthetic videos faithfully transferring the human time-varying head attributes from the source to the target actor. Our proposed implementation: 1) works in real time ($\sim 20$ fps), 2) runs on a commodity laptop with a webcam as the only input, 3) is interactive, allowing the participant to drive a target person, e.g. a celebrity, politician, etc, instantly by varying their expressions, head pose, and eye gaze, and visualising the synthesised video concurrently.
Deep neural networks achieve remarkable performances on a wide range of tasks with the aid of large-scale labeled datasets. However, large-scale annotations are time-consuming and labor-exhaustive to obtain on realistic tasks. To mitigate the requirement for labeled data, self-training is widely used in both academia and industry by pseudo labeling on readily-available unlabeled data. Despite its popularity, pseudo labeling is well-believed to be unreliable and often leads to training instability. Our experimental studies further reveal that the performance of self-training is biased due to data sampling, pre-trained models, and training strategies, especially the inappropriate utilization of pseudo labels. To this end, we propose Debiased, in which the generation and utilization of pseudo labels are decoupled by two independent heads. To further improve the quality of pseudo labels, we introduce a worst-case estimation of pseudo labeling and seamlessly optimize the representations to avoid the worst-case. Extensive experiments justify that the proposed Debiased not only yields an average improvement of $14.4$\% against state-of-the-art algorithms on $11$ tasks (covering generic object recognition, fine-grained object recognition, texture classification, and scene classification) but also helps stabilize training and balance performance across classes.
Assuming distributions are Gaussian often facilitates computations that are otherwise intractable. We study the performance of an agent that attains a bounded information ratio with respect to a bandit environment with a Gaussian prior distribution and a Gaussian likelihood function when applied instead to a Bernoulli bandit. Relative to an information-theoretic bound on the Bayesian regret the agent would incur when interacting with the Gaussian bandit, we bound the increase in regret when the agent interacts with the Bernoulli bandit. If the Gaussian prior distribution and likelihood function are sufficiently diffuse, this increase grows at a rate which is at most linear in the square-root of the time horizon, and thus the per-timestep increase vanishes. Our results formalize the folklore that so-called Bayesian agents remain effective when instantiated with diffuse misspecified distributions.
Differentiable simulators promise faster computation time for reinforcement learning by replacing zeroth-order gradient estimates of a stochastic objective with an estimate based on first-order gradients. However, it is yet unclear what factors decide the performance of the two estimators on complex landscapes that involve long-horizon planning and control on physical systems, despite the crucial relevance of this question for the utility of differentiable simulators. We show that characteristics of certain physical systems, such as stiffness or discontinuities, may compromise the efficacy of the first-order estimator, and analyze this phenomenon through the lens of bias and variance. We additionally propose an $\alpha$-order gradient estimator, with $\alpha \in [0,1]$, which correctly utilizes exact gradients to combine the efficiency of first-order estimates with the robustness of zero-order methods. We demonstrate the pitfalls of traditional estimators and the advantages of the $\alpha$-order estimator on some numerical examples.