Mehrnaz Amjadi

Deokwoo Jung, Nandini Ramanan, Mehrnaz Amjadi et al.

Anomaly detection for time-series data becomes an essential task for many data-driven applications fueled with an abundance of data and out-of-the-box machine-learning algorithms. In many real-world settings, developing a reliable anomaly model is highly challenging due to insufficient anomaly labels and the prohibitively expensive cost of obtaining anomaly examples. It imposes a significant bottleneck to evaluate model quality for model selection and parameter tuning reliably. As a result, many existing anomaly detection algorithms fail to show their promised performance after deployment. In this paper, we propose LaF-AD, a novel anomaly detection algorithm with label-free model selection for unlabeled times-series data. Our proposed algorithm performs a fully unsupervised ensemble learning across a large number of candidate parametric models. We develop a model variance metric that quantifies the sensitivity of anomaly probability with a bootstrapping method. Then it makes a collective decision for anomaly events by model learners using the model variance. Our algorithm is easily parallelizable, more robust for ill-conditioned and seasonal data, and highly scalable for a large number of anomaly models. We evaluate our algorithm against other state-of-the-art methods on a synthetic domain and a benchmark public data set.

Sankeerth Rao Karingula, Nandini Ramanan, Rasool Tahmasbi et al.

Time series forecasting is a fundamental task emerging from diverse data-driven applications. Many advanced autoregressive methods such as ARIMA were used to develop forecasting models. Recently, deep learning based methods such as DeepAr, NeuralProphet, Seq2Seq have been explored for time series forecasting problem. In this paper, we propose a novel time series forecast model, DeepGB. We formulate and implement a variant of Gradient boosting wherein the weak learners are DNNs whose weights are incrementally found in a greedy manner over iterations. In particular, we develop a new embedding architecture that improves the performance of many deep learning models on time series using Gradient boosting variant. We demonstrate that our model outperforms existing comparable state-of-the-art models using real-world sensor data and public dataset.

Mehrnaz Amjadi, Seyed Danial Mohseni Taheri, Theja Tulabandhula

Sequential recommendation systems model dynamic preferences of users based on their historical interactions with platforms. Despite recent progress, modeling short-term and long-term behavior of users in such systems is nontrivial and challenging. To address this, we present a solution enhanced by a knowledge graph called KATRec (Knowledge Aware aTtentive sequential Recommendations). KATRec learns the short and long-term interests of users by modeling their sequence of interacted items and leveraging pre-existing side information through a knowledge graph attention network. Our novel knowledge graph-enhanced sequential recommender contains item multi-relations at the entity-level and users' dynamic sequences at the item-level. KATRec improves item representation learning by considering higher-order connections and incorporating them in user preference representation while recommending the next item. Experiments on three public datasets show that KATRec outperforms state-of-the-art recommendation models and demonstrates the importance of modeling both temporal and side information to achieve high-quality recommendations.

Mehrnaz Amjadi, Theja Tulabandhula

Although the computational and statistical trade-off for modeling single graphs, for instance, using block models is relatively well understood, extending such results to sequences of graphs has proven to be difficult. In this work, we take a step in this direction by proposing two models for graph sequences that capture: (a) link persistence between nodes across time, and (b) community persistence of each node across time. In the first model, we assume that the latent community of each node does not change over time, and in the second model we relax this assumption suitably. For both of these proposed models, we provide statistically and computationally efficient inference algorithms, whose unique feature is that they leverage community detection methods that work on single graphs. We also provide experimental results validating the suitability of our models and methods on synthetic and real instances.