This position is offered through the ANU Computing Internship ([COMP4820] / [COMP8830])
Company
ANU Research School of Biology https://biology.anu.edu.au/
The ANU Research School of Biology (RSB) carries out research in a wide range of biological and biomedical sciences. It also oversees the University’s Biology curriculum, with a diverse offering of courses across both undergraduate and graduate degrees.
Project
Developing Enhanced Causal Counterfactual Explanations for Predictive Models
This research project is centered on the development of innovative algorithms for causal counterfactual explanations, a critical component in the interpretability of predictive models. The primary objective is to enhance the clarity and effectiveness of explanations provided by complex models. The project will involve:
Algorithm Design: Developing algorithms that generate insightful causal counterfactual explanations, which will help in understanding the specific changes needed in input features to achieve a desired change in the output. This involves creating sophisticated methods that leverage both traditional counterfactual reasoning and principles of causal inference to offer deeper insights into model behavior.
Validation through Experiments: Implementing these algorithms on diverse datasets, particularly those involving high-stakes decisions in sectors like finance and healthcare, to validate their practical utility and effectiveness. This stage will test the adaptability of the proposed methods across different types of predictive models and their capacity to improve decision-making processes.
Required technical skills
Required: Strong proficiency in Python, with extensive experience in data science and machine learning libraries.
Preferred: Hands-on experience with advanced machine learning frameworks like TensorFlow or PyTorch and familiarity with causal inference techniques, particularly the computational aspects.
Required/preferred professional and other skills
Required: Excellent analytical and conceptual thinking skills, with a proven ability to conduct independent research and algorithm development.
Preferred: Previous experience in quantitative research or algorithmic development in causal inference.
Delivery Mode
• Remote (Intern engages on project in a remote capacity)
Type of internship
Unpaid internship.
How to apply
Applications are invited from students who have already passed the eligibility checks for the Computing Internship courses COMP4820 or COMP8830. Further information about the Computing Internships can be found on the Computing Internship page.
You can nominate multiple preferred Internship projects and host organisations through the one application form.
Student applications will be open from the 13th to 29th Sep, 2024.