This week we have Brigitte and Ziwen presenting at our meeting
Agenda:
Brigitte’s talk:
“The title is: Using Boosted Regression Trees and remotely sensed data to drive decision-making
Abstract: Analysing Big Data is challenging due to the way they are stored, maintained and processed and recorded. In this presentation, I will demonstrate that BRTs can process different data granularities, heterogeneous data and in general data showing Big Data properties with satisfying results. We found in using a flexible, hierarchical, multivariate statistical machine learning algorithm, namely Boosted Regression Tree (BRT), there is a remarkable capability of addressing Big Data challenges.”
Ziwen’s talk:
“Title: Accelerating Bayesian Synthetic Likelihood with the Graphical Lasso
Description: As a simulation-based method, approximate Bayesian computation (ABC) can be used for problems where model simulation is available, without the evaluation requirement of the likelihood function. When the model summary statistic is roughly normally distributed in regions of the parameter space of interest, Bayesian synthetic likelihood (BSL), which uses a normal likelihood approximation for a summary statistic, is a useful method known to be more computationally efficient than ABC. In this presentation, I will show how to increase the computational efficiency of BSL even further by using the graphical lasso.”
We look forward to seeing you there!
Best regards,
Shovanur and Matt
Details:
| Location: | GP-Y801 |
| Start Date: | 17/08/2017 [add to calendar] |
| Start Time: | 13:00 |
| End Date: | 17/08/2017 |
| End Time: | 14:00 |
| Cost: | Nil |
| Organiser: | BRAG |