Université de Montréal

Mila - AI Ethics and Responsible AI Research

Canada · PhD/Postdoc

Mila is one of the world's largest academic AI research centers, founded by Yoshua Bengio. In AI ethics and responsible AI research, Mila deeply integrates technological advancement with ethical considerations, studying AI bias, fairness, interpretability, and social responsibility. It is a representative institution at the intersection of technology and ethics in AI alignment research.

学术历史

Mila was founded in 1993 (originally as the LISA laboratory) by Yoshua Bengio, and was renamed Mila in 2017, expanding into the Montreal AI research center. Bengio received the 2018 Turing Award for his contributions to deep learning. In the late 2010s, Bengio shifted his research focus toward AI safety and responsible AI, publishing multiple manifestos on AI safety and governance. Mila's unique positioning in AI ethics lies in having top-tier ML researchers lead ethical research, ensuring that ethical considerations are tightly coupled with technical implementation. In 2018, Bengio co-initiated the Montreal Declaration for Responsible AI, which became an important international consensus document in the field of AI ethics.

当前状态

Mila currently operates AI ethics and responsible AI research groups, with research directions encompassing algorithmic fairness, AI bias detection and mitigation, explainable AI, AI social impact assessment, and responsible AI development processes. Mila has a large-scale team of over 1,000 researchers, with AI ethics research embedded across all technical research directions. It hosts AI Safety-themed seminars and reading groups. Mila maintains collaborative relationships with companies including Google, Microsoft, and Samsung, while preserving academic independence.

实验室 / 研究中心

Mila - Quebec AI Institute

访问页面 →

关键人物

Yoshua Bengio

Professor / Founder of Mila

Deep learning pioneer and 2018 Turing Award recipient. In recent years, he has shifted his research focus toward AI safety and alignment, advocating for causal learning, System 2 cognition, and responsible AI development. Co-initiator of the Montreal Declaration for Responsible AI.

Golnoosh Farnadi

Assistant Professor

Expert in AI fairness and privacy research. Studies algorithmic bias detection, fair machine learning, and privacy-preserving ML, translating ethical principles into computable ML constraints.

Siva Reddy

Assistant Professor

Expert in natural language processing and AI safety research. Studies the societal impact, bias, and safety evaluation of large language models, bridging NLP technology with ethical analysis.

标志性成果

Yoshua Bengio's transition from deep learning pioneer to AI safety researcher, bringing a top-tier ML technical perspective to the alignment field
Montreal Declaration for Responsible AI (2018): an international consensus document in AI ethics defining ten principles for responsible AI
Technical implementation of algorithmic fairness: methodologies for embedding fairness constraints into ML training pipelines
Causal representation learning research: providing technical pathways for AI systems to understand causal relationships and human intent