FW26, William Gates Building
Unmarked review. Patterns across the cohort, not individual grades.
The CGF was \(\psi(\boldsymbol{\theta})\). After the break, \(A\) is action.
Success is low entropy in the reguland — with two caveats.
Quantity bound, not the structure theorem.
Name the existence claim. Fence the slogan.
Week 5: relative entropy needs a reference. Goals need one too.
The map recovers reward for a reward channel. It does not license general intelligence.
A channel is a conditional \(p(y\mid x)\). Rate cannot beat capacity.
Four results we will need later — stated, not proved today.
If \(X\to Y\to Z\) is Markov, processing cannot create information about \(X\).
If \(X\to Y\to Z\) is a Markov chain, processing cannot create information about \(X\).
Under Markov dynamics, \(\text{KL}\left( p_t\,\|\,\pi \right)\) falls — coarse-graining in time.
\[ \text{KL}\left( p\,\|\,\pi \right)=\text{KL}\left( w\,\|\,w_\pi \right)+\sum_i w_i\, \text{KL}\left( p_i\,\|\,\pi_i \right) \]
How much of equilibration is between groups, how much within?
Least-committal \(T\) given a relevance constraint — MaxEnt, now on mutual information.
Why “Inaccessible?”
From BLF axioms: Info gained = \(H(p) - H(q)\)
Our axiom: \(\sum h_i = C\) (constant)
\[\Delta(\sum h_i) = 0 \Rightarrow \text{observer learns nothing!}\]
Game Characteristics:
Physical Connections:
the Good Regulator Theorem of Conant and Ashby (1970)
requisite variety; Shannon form of Ashby (1956)
the Good Regulator Theorem of Conant and Ashby (1970)
requisite variety of Ashby (1956)
noisy channel coding / capacity of Shannon (1948)
Chapter 7 of Cover and Thomas (1991)
Theorem 2.8.1 of Cover and Thomas (1991)
the method; optional of Tishby et al. (1999)