Hear about the latest research in Psychology.

Title: Exploring signal detection models of recognition memory

Abstract: Computational models are frequently applied to behavioral data to summarize or describe the data. In the case of recognition memory, the unequal-variance Gaussian signal detection model is arguably the most popular mathematical model. While it was seen as merely mathematically convenient when first applied to recognition memory, it is now often discussed in terms of the theoretical inferences it allows about the latent processes underlying the observed behavior. In two related projects, we explore alternatives to the U-V Gaussian model, and discuss the implications for theoretical inference.

Calendar invitation circulated by Internal Seminar Coordinator: Suzanne Aussems with TEAMS link

We are pleased to announce that we have two excellent talks lined up for the upcoming March Warwick R User Group meeting! This is a joint session with the Birmingham R User Group, and will take place on Wednesday, 9th March, 5-6pm GMT. 

Meeting link: Click here to join the meeting

Meetup Event: https://www.meetup.com/WarwickRUG/events/284352459/

Talk 1 by Adnan Fiaz and Ben Dickel, Senior Data Scientists at National Grid

Title: Long term gas supply & demand scenarios at National Grid

Abstract: At National Grid (NG) our goal is to keep the lights on and homes warm. To achieve this goal we have to make sure our infrastructure is prepared for any future changes in demand. A high level future set of scenarios is provided to us by the Electricity System Operator (ESO). In this talk we will present how we go from this high level scenario to a much more detailed probabilistic outlook, one that allows us to assess the impact of changes on the network. Where relevant we will present examples of code but the talk will mostly be about our experiences and lessons learned.

Talk 2 by Dr Manuele Leonelli, Assistant Professor at School of Human Sciences and Technology, IE University 

Title: Sensitivity and robustness analysis in Bayesian networks with the bnmonitor R package

Abstract: Bayesian networks are a class of models that are widely used for risk assessment of complex operational systems. There are now multiple approaches, as well as implemented software, that guide their construction via data learning or expert elicitation. However, a constructed Bayesian network needs to be validated before it can be used for practical risk assessment. In this talk I will illustrate the usage of the bnmonitor R package: the first comprehensive software for the validation of a Bayesian network. An applied data analysis using bnmonitor is carried out over a medical dataset to illustrate the use of its wide array of functions. This is joint work with Ramsiya Ramanathan and Rachel Wilkerson.

Everyone is very welcome to join the meeting. If you’re interested in speaking at a future WRUG meeting or would just like to find out more about the group, please do get in touch.

Best wishes,

Aditi Shenvi 

(On behalf of the WRUG organising team) 

Dr Aditi Shenvi | Research Fellow at the Applied Statistics and Risk Unit

Department of Statistics

University of Warwick

aditi.shenvi@warwick.ac.uk | https://ashenvi10.github.io/