Time series are ubiquitous in many classification/regression applications. However, the time series data in real applications may contain many missing values. Hence, given multiple (possibly correlated) time series data, it is important to fill in missing values and at the same time to predict their class labels. Existing imputation methods often impose strong assumptions of the underlying data generating process, such as linear dynamics in the state space. In this paper, we propose a novel method, called BRITS, based on recurrent neural networks for missing value imputation in time series data.

Time series are ubiquitous in many classification/regression applications. However, the time series data in real applications may contain many missing values. Hence, given multiple (possibly correlated) time series data, it is important to fill in missing values and at the same time to predict their class labels. Existing imputation methods often impose strong assumptions of the underlying data generating process, such as linear dynamics in the state space. In this paper, we propose a novel method, called BRITS, based on recurrent neural networks for missing value imputation in time series data.

Neural Information Processing Systems http://nips.cc/

Neural models for text generation require a softmax layer with proper word embeddings during the decoding phase. Most existing approaches adopt single point embedding for each word. However, a word may have multiple senses according to different context, some of which might be distinct. In this paper, we propose KerBS, a novel approach for learning better embeddings for text generation. KerBS embodies two advantages: a) it employs a Bayesian composition of embeddings for words with multiple senses; b) it is adaptive to semantic variances of words and robust to rare sentence context by imposing learned kernels to capture the closeness of words (senses) in the embedding space. Empirical studies show that KerBS significantly boosts the performance of several text generation tasks.

This problem is closely related to domain adaptation, and in our case, is motivated by the need to combine clinical and imaging based biomarkers from multiple sites and/or batches - a fairly common impediment in conducting analyses with much larger sample sizes. We address this problem using ideas from hypothesis testing on the transformed measurements, wherein the distortions need to be estimated in tandem with the testing. We derive a simple algorithm and study its convergence and consistency properties in detail, and provide lower-bound strategies based on recent work in continuous optimization. On a dataset of individuals at risk for Alzheimer's disease, our framework is competitive with alternative procedures that are twice as expensive and in some cases operationally infeasible to implement.