The Computational Media Lab at the ANU focus on computational problems for understanding online media and their interactions with and among humans. We develop core methods in machine learning and optimization, and use them to formulate and solve problems in multimedia undestanding and online behavior analysis.
This post outlines techniques for computing the expected event rate for Hawkes processes, or the so-called Hawkes Intensity Process (HIP).
We give a brief overview to a new method for computing expected event rate in unit time for point processes. This is important for estimating interval-censored Hawkes processes – or when the volume of events, and not individual event times are known.Read More
We discuss three approaches for graph inversion from steady state distribution of random walks. This first post compares the problem settings.
Markov chains and random walk on graphs are long-standing subjects in applied math and related disciplines. Given a markov chain / transition matrix / graph, properties of the random walk such as steady state distribution, mixing time are well-studied. Only in recent years, however, the inverse problem has started gaining interest – that is, given some measurements after random walk on a graph, infer some version of the graph. This blog post compares three related but different problem settings in recent literature from
We explore Australian caves and cherries in the second half of 2016.
Australia has great weather for the ourdoors, no matter it’s winter or summer – that’s what we did, and here’s photo proof.Read More
We gave presentations at ACM CIKM and the D2D CRC Annual Conference.
Our work received some pulicity in the “fall” conference season internationally and in Australia.Read More
We pbridge the gap between problem settings and approaches for predicting the size of a retweeting cascade.
Predicting popularity as number of retweets a tweet will get is an important and difficult task. It’s unclear which approaches, settings and features works best. Our current CIKM ‘16 paper bridges this gap by comparing across feature driven and point process approaches under both regression and classification settings.
We present a novel approach to jointly rank points-of-interest and plan routes
The problem of recommending tours to travellers is an important and broadly studied area. We consider the task of recommending a sequence of points-of-interest (POI), that simultaneously uses information about POIs and routes.
We propose a new probabilistic knowledge base factorisation that benefits from the path structure of existing knowledge.
Relational knowledge graphs formalise our understanding about the world and help us reason and infer in a wide range of tasks. The construction of a knowledge graph is an active research area with many important and challenging research questions. Throughout this research, we address some important problems in the knowledge graph construction and propose novel statistical relational models to solve the problems.Read More
We plot the citation behavior over time for different subfields in computer science, using data from microsoft academic graph.
We plot the citation survival rate of many computer science conferences, using data from microsoft academic graph.
Congratulations to all authors!
We have great news! Three papers have been accepted to CIKM 2016.
“Feature Driven and Point Process Approaches for Popularity Prediction” - Swapnil Mishra, Marian-Andrei Rizoiu, and Lexing Xie
“Learning Ranks and Routes to Recommend Trajectories” - Dawei Chen, Lexing Xie, and Cheng Soon Ong
“Probabilistic Knowledge Graph Construction: Compositional and Incremental Approaches” - Dongwoo Kim, Lexing Xie, and Cheng Soon Ong
We plot the incoming and outgoing citation flow of many computer science conferences, using data from microsoft academic graph.
Our research is broadly concerned with automated undertanding of rich media, esp. pictures and language; and modeling of collective online behaviours. We specialize and innovate in different method in machine leanring and optimization, recent favorites include: stochastic time series models, sequence and language models with neural netoworks, matrix and tensor factorisation, active learning, structured prediction models.Read More