Amir-Hossein Karimi
Bio:
Dr. Amir-Hossein Karimi is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of Waterloo, and a Vector Institute Faculty Affiliate. He leads the Collaborative Human-AI Reasoning Machines (CHARM) Lab, dedicated to pioneering advances in artificial intelligence to facilitate trustworthy human-AI collaborations. The CHARM Lab's research agenda includes developing AI systems that can ensure AI safety and ethics, rectify adverse outcomes, and synergize human and machine capabilities, particularly in domains like healthcare and education. This involves research and applied work causal inference, explainable AI, and program synthesis.
Dr. Karimi's scholarly contributions have been showcased at esteemed AI and machine learning venues including NeurIPS, ICML, AAAI, AISTATS, ACM-FAccT, and ACM-AIES, and he has authored influential publications like a comprehensive survey paper in the prestigious ACM Computing Surveys. Prior to joining UWaterloo, he accumulated significant industry experience at leading tech companies such as BlackBerry, Meta, Google Brain, and DeepMind, and consulted for various startups and incubators. His academic and professional achievements have earned him multiple accolades, such as the Spirit of Engineering Science Award (University of Toronto, 2015), the Alumni Gold Medal Award (University of Waterloo, 2018), the NSERC Canada Graduate Scholarship (2018), the Google PhD Fellowship (2021), the ETH Zurich Medal (2024), and the Igor Ivkovic Teaching Excellence Award (2024).
Dr. Karimi is a firm believer in the strength of diverse perspectives and is dedicated to building inclusive spaces in academia and beyond. As a professor, he strives to inspire the next generation of AI researchers and engineers, helping them understand and responsibly apply the power of AI. Further emphasizing his commitment to inclusive learning, he co-founded "PrinceOfAI," an initiative that provides free education on basic and advanced AI topics to a community of over 30,000 individuals. This initiative aims to reduce barriers to education and offer opportunities to those who traditionally lack access. With a wealth of content equivalent to an introductory Machine Learning course, their platform facilitates active learning through technical posts, engaging video reels, and interactive webinars.
Joining our team:
I am always on the lookout for exceptional and highly motivated students/visitors across all levels (Bachelor's, Master's, PhD, Post-doctoral). If you're passionate about building the future of trustworthy human-AI symbiosis, please fill out this form.
News:
2024.04: "Prospector Heads: Generalized Feature Attribution for Large Models & Data" accepted to ICML.
2024.03: Excited about receiving a NSERC Discovery Grant; more research coming soon!
2024.03: Very honoured to have been nominated by my students and selected for Igor Ivkovic Teaching Excellence Award during my first teaching term.
2023.11: Honoured that my PhD dissertation was awarded the ETH Zurich Medal
2023.09: Started my new role as an assistant professor in machine learning 🤖🧠 at the University of Waterloo and the Vector Institute
2023.05: "Robustness Implies Fairness in Causal Algorithmic Recourse" accepted at FAccT 2023.
2023.04: Two papers accepted to ICML 2023.
2023.03: Workshop on Counterfactuals in Minds & Machines accepted to ICML.
2022.05 - 2022.10: Had a wonderful internship at DeepMind in London, UK.
2022.05: "On the Robustness of Causal Algorithmic Recourse" accepted to ICML.
2022.03: "A survey of algorithmic recourse" accepted to ACM Computing Surveys.
2022.02: Co-instructed "Causethical ML" graduate seminar at Saarland University.
2022.02: "On the fairness of causal algorithmic recourse" accepted at AAAI (Oral).
2021.12 - 2022.04: Had a great time at an internship at Google Brain in Canada.
Research Focus:
In this AI era, there's an urgent need to advance responsible, ethical AI systems in response to global needs. We must work to gain trust in AI, especially in algorithmic recourse and creating reliable human-machine collaborations. Such tasks form the core of ongoing AI research challenges, primarily harmonizing technological progress with societal approval. Our lab's research aims to contribute to ethical AI research and addresses key issues through several goals:
1. Develop methods for algorithmic recourse to recover from or overturn unfavourable AI-driven decisions. This involves managing data uncertainties, addressing diverse stakeholder needs, and ensuring real-world recourse applications.
2. Employ causal-based methods to build safe, truthful, and ethical AI systems, aiming to boost transparency and trust in AI decisions and explanations thereof. This strategy will expose biases, encourage ethical practices, and earn public AI trust.
3. Develop hybrid human-AI systems for industries such as healthcare, science, and education, to foster trustworthy, responsible, and human-centric AI solutions.
We plan an integrated approach in this research, using explainable AI for transparency, program synthesis for expanding user interaction with automated systems, and causal inference for managing uncertainties. This integrated approach will enhance our focus on human-centric AI, crucial for widespread sector adoption.
Feedback:
Always happy to hear feedback/comments/suggestions!
Nasir-ol-molk Mosque, Shiraz, 2019
Alumni Gold Medal Award, UWaterloo, 2018
Paris, 2019