I am a post-doctoral researcher at Carnegie Mellon University and Bosch Center for AI, working with Prof. Zico Kolter. My research focuses on developing new frameworks for learning algorithm design that align with the ultimate goal of the learning system deployment. This capability is lacking for most of current machine learning models. Aligning the learning algorithms with the ultimate goal in which the algorithms are deployed is an important milestone to unlock the full potential of machine learning to solve real-world problems.

Specifically, I developed a framework for designing learning algorithm that align with the goal of optimizing performance metrics in differentiable learning pipelines. I investigate for new paradigms in designing learning algorithms that align with various types of goals commonly required by machine learning applications. Some of keywords related to my research areas are: goal-oriented learning, evaluation metrics, differentiable learning, structured prediction, graphical models, and fairness in ML.

I received my Ph.D. from the Department of Computer Science, University of Illinois at Chicago. I was fortunate to have Prof. Brian Ziebart as my advisor. I was grateful to also collaborate with Prof. Xinhua Zhang.

I was born and raised in a small village in Java island, Indonesia. After high school, I moved to Jakarta to complete my bachelor degree at the Institute of Statistics. I was also very lucky to be awarded Fulbright scholarship in 2012. Being a Fulbright grantee was a milestone to pursue my passion in machine learning research.

I am fond of nature. In my spare time, I enjoy hiking in national parks and local forests with my family.

Selected Publications: [All Publications]

  1. AP-Perf: Incorporating Generic Performance Metrics in Differentiable Learning Rizal Fathony, and J. Zico Kolter Conference: International Conference on Artificial Intelligence and Statistics (AISTATS) 2020 Link PDF Code (Julia) Code (Python)
  2. Distributionally Robust Graphical Models Rizal Fathony, Ashkan Rezaei, Mohammad Ali Bashiri, Xinhua Zhang, and Brian Ziebart Conference: Advances in Neural Information Processing Systems (NeurIPS) 2018 Link PDF Poster Code
  3. Consistent Robust Adversarial Prediction for General Multiclass Classification Rizal Fathony, Kaiser Asif, Anqi Liu, Mohammad Ali Bashiri, Wei Xing, Sima Behpour, Xinhua Zhang, and Brian D Ziebart Preprint: arXiv preprint 2018 Link
  4. Adversarial Multiclass Classification: A Risk Minimization Perspective Rizal Fathony, Anqi Liu, Kaiser Asif, and Brian Ziebart Conference: Advances in Neural Information Processing Systems (NeurIPS) 2016 Link PDF Poster Code