• Publications
  • Influence
Spamming botnets: signatures and characteristics
An in-depth analysis of the identified botnets revealed several interesting findings regarding the degree of email obfuscation, properties of botnet IP addresses, sending patterns, and their correlation with network scanning traffic. Expand
Inductive Representation Learning on Temporal Graphs
The temporal graph attention (TGAT) layer is proposed to efficiently aggregate temporal-topological neighborhood features as well as to learn the time-feature interactions by developing a novel functional time encoding technique based on the classical Bochner's theorem from harmonic analysis. Expand
How dynamic are IP addresses?
This paper introduces a novel algorithm, UDmap, to identify dynamically assigned IP addresses and analyze their dynamics pattern, which is the first successful attempt to automatically identify and understand IP address dynamics. Expand
Using the wisdom of the crowds for keyword generation
This work identifies queries related to a campaign by exploiting the associations between queries and URLs as they are captured by the user's clicks, and proposes algorithms within the Markov Random Field model to solve this problem. Expand
Robust variational speech separation using fewer microphones than speakers
A variational inference algorithm for robust speech separation, capable of recovering the underlying speech sources even in the case of more sources than microphone observations, is presented. TheExpand
Generative Graph Convolutional Network for Growing Graphs
A unified generative graph convolutional network that learns node representations for all nodes adaptively in a generative model framework by sampling graph generation sequences constructed from observed graph data is proposed. Expand
Unsupervised image translation
This work describes a computationally efficient, probabilistic inference and learning algorithm for inferring the most likely output image and learning the rendering style and demonstrates the technique using several tasks, including rendering a photograph in the artistic style of an unrelated scene, de-noising, and texture transfer. Expand
Product Knowledge Graph Embedding for E-commerce
The key entities are defined and summarize the pivotal product relations that are critical for general e-commerce applications including marketing, advertisement, search ranking and recommendation. Expand
Self-attention with Functional Time Representation Learning
This work introduces a functional feature map that embeds time span into high-dimensional spaces and reveals the functional forms of the feature map under classic functional function analysis results, namely Bochner's Theorem and Mercer’s Theorem. Expand
Probabilistic Inference of Speech Signals from Phaseless Spectrograms
A generative model of time-domain speech signals and their spectrograms is described, and it is shown how an efficient optimizer can be used to find the maximum a posteriori speech signal, given the spectrogram. Expand