Aligning AI With Shared Human Values (ICLR 2021)
-
Updated
Apr 21, 2023 - Python
Aligning AI With Shared Human Values (ICLR 2021)
List of references about Machine Learning bias and ethics
[AAAI 2018] Implementation of the Ethics Shaping approach proposed in "A low-cost ethics shaping approach for designing reinforcement learning agents"
Code and data for Paper "Enhancing Ethical Explanations of Large Language Models through Iterative Symbolic Refinement"
[work-in-progress] Curated list of standards related to Ethics of Autonomous and Intelligent Systems (A/IS)
[work-in-progress] Curated list of organizations related to Ethics in Artificial Intelligence and Autonomous Systems (AI/AS)
Reinforcement learning environment for learning ethical behaviours in a SmartGrid use-case.
Probabilistic Moral Planner based on heuristic Dynamic Programming AO* and Machine Ethics Hypothetical Retrospection argumentation. Works with conflicting moral theories and non-moral costs/goals.
Value aligned socio-political-economic systems
A Julia value-theory toolkit for making normative criteria explicit in computational systems, with typed value concepts and operations to assess, optimize and reason about value-aligned behaviour in AI.
Seeding mercy and coexistence - Socratic Method Dia-LOGs for LLM Alignment
AI-HPP-Standard: an inspection-ready architecture for accountable AI systems. Vendor-neutral. Audit-ready. High-risk gated. Developed via structured multi-model orchestration with human oversight. Designed to support emerging international AI governance.
Graduate course 11118BLG001 (SDÜ): Machine Ethics and AI Safety. 14 weeks of lecture notes, interactive browser labs, Colab notebooks, knowledge checks and capstone projects on fairness, alignment, robustness, interpretability, LLM security and AI governance. Open for reuse (CC BY 4.0, MIT).
Code used for my master's thesis - Artificial Morality: Incorporating Moral Work into Artificial Intelligence
Theory of Freedom — book and philosopher edition for AI morals. A1–A7 legitimacy floor. CC BY 4.0.
Uses Montague semantics and a deterministic graph database to enforce mathematically verified, immutable ethical logic, prevents utilitarian overrides of deontological constraints via an immutable constraint layer.
Mathematical Conscience Framework - 10 AIs collaborated.
[work-in-progress] Curated list of Open Groups (accessible to persons who are not yet experts and do not require association with an institution) to discuss Ethics of Autonomous and Intelligent Systems (A/IS)
To associate your repository with the machine-ethics topic, visit your repo's landing page and select "manage topics."