Algorithmic Fairness
Last edited: August 8, 2025Could algorithms make more equitable decisions? No: algorithms may propagate and amplify biases—its not enough just to learn/optimize.
Collaboration with other fields has both a language gap and a value gap.
Technical Algorithmic Fairness
fair scheduling, distributed computing, envy-freeness, cake cutting, stable matching, etc.
Interpreting Individual Probabilities
“I have n% probability of developing this thing” — what does this mean? what does it capture in the environment?
Applications
- insurance: how does actuarial scenarios work for this case—it is, in those cases, context dependent
- satisfying fairness maybe mis-generalized: discrimination can be subtle for a given measure (do you know it when you see it?)
Definitions of fairness, however, is important to characterize a system.
Ali Partovi
Last edited: August 8, 2025Kwotes
“Working with big tech is a job closer to big tech.”
Align with New Vocab
Last edited: August 8, 2025Begin with a new installation of MFA, and head to the directory. First run validate with the original dictionary.
mfa validate ~/Downloads/tb/my_corpus english_us_arpa english_us_arpa
We see that there is in deed an section of corpus that is out-of-vocab.
INFO - 11 OOV word types
INFO - 18 total OOV tokens
Therefore, we will generate a new dictionary based on the existing dictionary of english_us_arpa.
First download the english_us_arpa model
mfa model download g2p english_us_arpa
Then, perform the actual dictionary generation:
Alivio April Checkin
Last edited: August 8, 2025- Want to interview more severe ashma
- Want to find someone younger
Difference between marketing and purchaser.
Taking to people
Spoke with Matt. Talked with more details with prototyping and how they can build a unique product.
Have not gotten back to him yet.
alpha vector
Last edited: August 8, 2025Recall, from conditional plan evaluation, we had that:
\begin{equation} U^{\pi}(b) = \sum_{s}^{} b(s) U^{\pi}(s) \end{equation}
let’s write it as:
\begin{equation} U^{\pi}(b) = \sum_{s}^{} b(s) U^{\pi}(s) = {\alpha_{\pi}}^{\top} b \end{equation}
where \(\U_{\pi}(s)\) is the conditional plan evaluation starting at each of the initial states.
\begin{equation} \alpha_{\pi} = \qty[ U^{\pi}(s_1), U^{\pi}(s_2) ] \end{equation}
You will notice, then the utility of \(b\) is linear on \(b\) for different policies \(\alpha_{\pi}\):

At every belief \(b\), there is a policy which has the highest \(U(b)\) at that \(b\) given be the alpha vector formulation.
