Ana Lucic , PhD student at the information and language processing systems research group, ILPS, has won the best student paper (cs) at the annual FAT* conference in Barcelona last week.
The FAT* conference is a computer science conference with a cross-disciplinary focus that brings together researchers and practitioners interested in fairness, accountability, and transparency in socio-technical systems.
Ana’s paper is entitled "Why Does My Model Fail? Contrastive Local Explanations for Retail Forecasting.” Mistakes have a significant impact on users’ perception of predictive models. By focusing on automatically generating explanations of mistakes, the paper gives insights into the phenomenon of algorithm aversion while also giving users the types of explanations they are interested in seeing.