Alan Chan: Aligning AI with Human Values

Table of Contents

1. Introduction
2. Alan Chan's Background and Research Interests
3. Value Alignment and AI Governance
4. Measurement of Harms from Language Models
5. Incentives for Agents to Impact the World
6. Regulation of Explainability in Algorithmic Systems
7. Scoring Rules for Performative Binary Prediction
8. Effects of Global Exclusion in AI Development
9. Approaching Ethical Impacts in AI Research
10. Inverse Policy Evaluation for Value-Based Sequential Decision Making
11. Normal Accidents and AI Systems

Alan Chan: Aligning AI Systems with Human Values

Artificial intelligence (AI) has the potential to revolutionize the world as we know it. However, as with any powerful tool, there are risks associated with its use. One of the biggest concerns is the alignment of AI systems with human values. Alan Chan, a PhD student at Mila, the Montreal Institute for Learning Algorithms, is working to address this issue.

Alan Chan's Background and Research Interests

Before joining Mila, Alan was a master's student at the Alberta Machine Intelligence Institute and the University of Alberta, where he worked with Martha White. Alan's expertise and research interests encompass value alignment and AI governance. He is currently exploring the measurement of harms from language models and the incentives that agents have to impact the world. His projects have examined the regulation of explainability in algorithmic systems, scoring rules for performative binary prediction, the effects of global exclusion in AI development, and the role of a graduate student in approaching the ethical impacts in AI research. In addition, Alan has conducted research into inverse policy evaluation for value-based sequential decision making and the concept of normal accidents and AI systems.

Value Alignment and AI Governance

Alan's research is motivated by the need to Align AI systems with human values and his passion for scientific and governance work in this field. He believes that value alignment is crucial for ensuring that AI systems are beneficial to society. Alan's work in AI governance focuses on developing policies and regulations that promote the responsible use of AI.

Measurement of Harms from Language Models

One of Alan's Current projects involves the measurement of harms from language models. Language models are AI systems that can generate human-like text. However, they can also be used to spread misinformation and hate speech. Alan is working to develop methods for measuring the harm caused by language models and to develop strategies for mitigating this harm.

Incentives for Agents to Impact the World

Another area of Alan's research is the incentives that agents have to impact the world. AI systems are designed to achieve certain goals, but these goals may not align with human values. Alan is working to develop methods for aligning the goals of AI systems with human values and to ensure that AI systems are incentivized to act in ways that are beneficial to society.

Regulation of Explainability in Algorithmic Systems

Alan's work on the regulation of explainability in algorithmic systems focuses on developing methods for ensuring that AI systems are transparent and accountable. Explainability is crucial for ensuring that AI systems can be audited and that their decisions can be understood by humans.

Scoring Rules for Performative Binary Prediction

Performative binary prediction is a Type of AI system that is used to make decisions based on binary outcomes. Alan's work on scoring rules for performative binary prediction focuses on developing methods for ensuring that these systems are fair and unbiased.

Effects of Global Exclusion in AI Development

Alan's work on the effects of global exclusion in AI development focuses on the impact of AI on developing countries. He is working to ensure that AI development is inclusive and that the benefits of AI are shared by all.

Approaching Ethical Impacts in AI Research

Alan's work on approaching ethical impacts in AI research focuses on developing methods for ensuring that AI research is conducted in an ethical and responsible manner. He believes that it is crucial for researchers to consider the ethical implications of their work and to ensure that their research is aligned with human values.

Inverse Policy Evaluation for Value-Based Sequential Decision Making

Inverse policy evaluation is a type of AI system that is used to make decisions based on value-based sequential decision making. Alan's work on inverse policy evaluation focuses on developing methods for ensuring that these systems are aligned with human values and that they make decisions that are beneficial to society.

Normal Accidents and AI Systems

Alan's work on normal accidents and AI systems focuses on the potential risks associated with the use of AI. He believes that it is crucial for researchers to consider the potential risks of AI and to develop strategies for mitigating these risks.

In conclusion, Alan Chan's work is focused on aligning AI systems with human values and ensuring that AI is used in a responsible and ethical manner. His research is crucial for ensuring that the benefits of AI are shared by all and that the potential risks of AI are mitigated.