Preference Elicitation
Last edited: January 1, 2026For weighted sum method for instance, we need to figure a \(w\) such that:
\begin{equation} f = w^{\top}\mqty[f_1 \\ \dots\\f_{N}] \end{equation}
where weight \(w \in \triangle_{N}\).
To do this, we essentially infer the weighting scheme by asking “do you like system \(a\) or system \(b\)”.
- first, we collect a series of design variables \((a_1, a_2, a_3 …)\) and \((b_1, b_2, b_3…)\) and we ask “which one do you like better”
- say our user WLOG chose \(b\) over \(a\)
- so we want to design a \(w\) such that \(w^{\top} a < w^{\top} b\)
- meaning, we solve for a \(w\) such that…
\begin{align} \min_{w}&\ \sum_{i=1}^{n} (a_{i}-b_{i})w^{\top} \\ \text{such that}&\ \bold{1}^{\top} w = 1 \\ &\ w \geq 0 \end{align}
principles of biomedical ethics
Last edited: January 1, 2026- autonomy
- informed consent
- beneficence
- non-maleficence
- justice
Reliable RL
Last edited: January 1, 2026Thinking about advances in the capabilities of RL: Knowledge Discovery -> Reasoning (programming assistance) ->(ongoing)-> Robotics
Insight: as time goes on, the “risk-criticality” of our applications increase; yet, as risk critical scenarios increase, its harder to get data.
Reliable Feedback Loop
General desirable structure…
Verify (claims and requirements) => Safeguard (safe continuous deployment) => Generalize (via compositional generalization—incrementing adding behavior without loosing behavior) => Verify => …
Deal with Stochasticity
An RL algorithm is explicable, if, WHP, running on the same MDP with fixed randomness results in the same outcomes.
SU-EE364A JAN132026
Last edited: January 1, 2026Key Sequence
Notation
New Concepts
Important Results / Claims
- operations that preserve fuction convexity
- composition with scalar functions
- general composition rule that preserve convexity
- constructive convexity verification
Questions
Interesting Factoids
SU-PHIL2 APR012025
Last edited: January 1, 2026Challenge of moral philosophy: a system which resolves morality and ethics together.
tools
- definitions
- appearance vs. reality (descriptive vs. true values)
- reflective equilibrium
morality
morality is paradimatically a set of rules/expectations (concerning character/motives/emotions) for right/wrong behavior.
A code of conduct that people “must” follow to…
- regulate / guide interpersonal interactions
- rules that concern…
- harm / benefit
- justice / fairness
- loyalty / obedience
- sanctity / purity
Key question of morality: what do we owe to each other? what do I owe other people?
