Selected Student Projects in XAI

Explaining The Characteristics of The HVAC Systems Model

Han-Gyeong Chu, Jae-Hwan Jeon

As part of the Building Energy Management System (BEMS), the Heating, Ventilation, and Air Condition system (HVAC ) applies various technologies to control the temperature, humidity, and purity of the air in an enclosed space. It helps to manage energy and reduce consumption costs. To optimize the HVAC application, a central controller “Model Predictive Control (MPC)” is introduced to this entire system.

For example, we aim to have the room temperature reach 25 degrees at 9 am. To achieve this goal, MPC might start to operate from 7:30 am. Such a schedule is constantly altered regarding many external variables such as external temperature, humidity, etc. This project compares and discusses physics-based models, normal AI, and XAI to better predict a future temperature.


Presentation File

XAI_Project FP_Team2 Energy and Buildings_submit.pdf의 사본

Bias in Criminal Sentencing

Andreas Cho Woehler, Naomi Xu, Nishadi Prasangini

The sentencing decision in a criminal trial can have enormous individual and collective impacts. Either it can affect a defendant’s life in terms of job prospects and relationships with friends and family, or it can reduce the social gap among people and mediate the conflict between the authorities and a particular group.

However, as humans still make sentencing decisions, it is reasonable to expect some unconscious biases to influence the sentencing outcome. Knowing and understanding the preferences can help create more fair sentencing decisions. Aiming to Identify and recognize biases in sentencing decisions, this project determines factors that will influence sentence length and their magnitude via various prediction models and provide suggestions to solve bias issues.


Presentation File

Team5_NNA_ExtendedPresentation.pptx의 사본