$\require{mathtools}
%
%%% GENERIC MATH %%%
%
% Environments
\newcommand{\al}[1]{\begin{align}#1\end{align}} % need this for \tag{} to work
\renewcommand{\r}{\mathrm}
%
% Greek
\newcommand{\eps}{\epsilon}
\newcommand{\veps}{\varepsilon}
\newcommand{\Om}{\Omega}
\newcommand{\om}{\omega}
\newcommand{\Th}{\Theta}
\let\fi\phi % because it looks like an f
\let\phi\varphi % because it looks like a p
%
% Miscellaneous shortcuts
% .. over and under
\newcommand{\ss}[1]{_{\substack{#1}}}
\newcommand{\ob}{\overbrace}
\newcommand{\ub}{\underbrace}
\newcommand{\ol}{\overline}
\newcommand{\tld}{\widetilde}
\newcommand{\HAT}{\widehat}
\newcommand{\f}{\frac}
\newcommand{\s}[2]{#1 /\mathopen{}#2}
\newcommand{\rt}[1]{ {\sqrt{#1}}}
% .. relations
\newcommand{\sr}{\stackrel}
\newcommand{\sse}{\subseteq}
\newcommand{\ce}{\coloneqq}
\newcommand{\ec}{\eqqcolon}
\newcommand{\ap}{\approx}
\newcommand{\ls}{\lesssim}
\newcommand{\gs}{\gtrsim}
% .. miscer
\newcommand{\q}{\quad}
\newcommand{\qq}{\qquad}
\newcommand{\heart}{\heartsuit}
%
% Delimiters
% (I needed to create my own because the MathJax version of \DeclarePairedDelimiter doesn't have \mathopen{} and that messes up the spacing)
% .. one-part
\newcommand{\p}[1]{\mathopen{}\left( #1 \right)}
\newcommand{\b}[1]{\mathopen{}\left[ #1 \right]}
\newcommand{\set}[1]{\mathopen{}\left\{ #1 \right\}}
\newcommand{\abs}[1]{\mathopen{}\left\lvert #1 \right\rvert}
\newcommand{\floor}[1]{\mathopen{}\left\lfloor #1 \right\rfloor}
\newcommand{\ceil}[1]{\mathopen{}\left\lceil #1 \right\rceil}
\newcommand{\inner}[1]{\mathopen{}\left\langle #1 \right\rangle}
% .... (use phantom to force at least the standard height of double bars)
\newcommand{\norm}[1]{\mathopen{}\left\lVert #1 \vphantom{f} \right\rVert}
\newcommand{\frob}[1]{\norm{#1}_\mathrm{F}}
%% .. two-part
\newcommand{\incond}[2]{#1 \mathop{}\middle|\mathop{} #2}
\newcommand{\cond}[2]{ {\left.\incond{#1}{#2}\right.}}
\newcommand{\pco}[2]{\p{\incond{#1}{#2}}}
\newcommand{\bco}[2]{\b{\incond{#1}{#2}}}
\newcommand{\setco}[2]{\set{\incond{#1}{#2}}}
\newcommand{\at}[2]{ {\left.#1\right|_{#2}}}
\newcommand{\pat}[2]{\p{\at{#1}{#2}}}
\newcommand{\bat}[2]{\b{\at{#1}{#2}}}
% ..... (use phantom to force at least the standard height of double bar)
\newcommand{\oldpara}[2]{#1\vphantom{f} \mathop{}\middle\|\mathop{} #2}
%\newcommand{\para}[2]{#1\vphantom{f} \mathop{}\middle\|\mathop{} #2}
\newcommand{\para}[2]{\mathchoice{\begin{matrix}#1\\\hdashline#2\end{matrix}}{\begin{smallmatrix}#1\\\hdashline#2\end{smallmatrix}}{\begin{smallmatrix}#1\\\hdashline#2\end{smallmatrix}}{\begin{smallmatrix}#1\\\hdashline#2\end{smallmatrix}}}
\newcommand{\ppa}[2]{\p{\para{#1}{#2}}}
\newcommand{\bpa}[2]{\b{\para{#1}{#2}}}
%\newcommand{\bpaco}[4]{\bpa{\incond{#1}{#2}}{\incond{#3}{#4}}}
\newcommand{\bpaco}[4]{\bpa{\cond{#1}{#2}}{\cond{#3}{#4}}}
%
% Levels of closeness
\newcommand{\scirc}[1]{\sr{\circ}{#1}}
\newcommand{\sdot}[1]{\sr{.}{#1}}
\newcommand{\slog}[1]{\sr{\log}{#1}}
\newcommand{\createClosenessLevels}[7]{
\newcommand{#2}{\mathrel{(#1)}}
\newcommand{#3}{\mathrel{#1}}
\newcommand{#4}{\mathrel{#1\!\!#1}}
\newcommand{#5}{\mathrel{#1\!\!#1\!\!#1}}
\newcommand{#6}{\mathrel{(\sdot{#1})}}
\newcommand{#7}{\mathrel{(\slog{#1})}}
}
\let\lt\undefined
\let\gt\undefined
% .. vanilla versions (is it within a constant?)
\newcommand{\ez}{\scirc=}
\newcommand{\eq}{\simeq}
\newcommand{\eqq}{\mathrel{\eq\!\!\eq}}
\newcommand{\eqqq}{\mathrel{\eq\!\!\eq\!\!\eq}}
\newcommand{\lez}{\scirc\le}
\newcommand{\lq}{\preceq}
\newcommand{\lqq}{\mathrel{\lq\!\!\lq}}
\newcommand{\lqqq}{\mathrel{\lq\!\!\lq\!\!\lq}}
\newcommand{\gez}{\scirc\ge}
\newcommand{\gq}{\succeq}
\newcommand{\gqq}{\mathrel{\gq\!\!\gq}}
\newcommand{\gqqq}{\mathrel{\gq\!\!\gq\!\!\gq}}
\newcommand{\lz}{\scirc<}
\newcommand{\lt}{\prec}
\newcommand{\ltt}{\mathrel{\lt\!\!\lt}}
\newcommand{\lttt}{\mathrel{\lt\!\!\lt\!\!\lt}}
\newcommand{\gz}{\scirc>}
\newcommand{\gt}{\succ}
\newcommand{\gtt}{\mathrel{\gt\!\!\gt}}
\newcommand{\gttt}{\mathrel{\gt\!\!\gt\!\!\gt}}
% .. dotted versions (is it equal in the limit?)
\newcommand{\ed}{\sdot=}
\newcommand{\eqd}{\sdot\eq}
\newcommand{\eqqd}{\sdot\eqq}
\newcommand{\eqqqd}{\sdot\eqqq}
\newcommand{\led}{\sdot\le}
\newcommand{\lqd}{\sdot\lq}
\newcommand{\lqqd}{\sdot\lqq}
\newcommand{\lqqqd}{\sdot\lqqq}
\newcommand{\ged}{\sdot\ge}
\newcommand{\gqd}{\sdot\gq}
\newcommand{\gqqd}{\sdot\gqq}
\newcommand{\gqqqd}{\sdot\gqqq}
\newcommand{\ld}{\sdot<}
\newcommand{\ltd}{\sdot\lt}
\newcommand{\lttd}{\sdot\ltt}
\newcommand{\ltttd}{\sdot\lttt}
\newcommand{\gd}{\sdot>}
\newcommand{\gtd}{\sdot\gt}
\newcommand{\gttd}{\sdot\gtt}
\newcommand{\gtttd}{\sdot\gttt}
% .. log versions (is it equal up to log?)
\newcommand{\elog}{\slog=}
\newcommand{\eqlog}{\slog\eq}
\newcommand{\eqqlog}{\slog\eqq}
\newcommand{\eqqqlog}{\slog\eqqq}
\newcommand{\lelog}{\slog\le}
\newcommand{\lqlog}{\slog\lq}
\newcommand{\lqqlog}{\slog\lqq}
\newcommand{\lqqqlog}{\slog\lqqq}
\newcommand{\gelog}{\slog\ge}
\newcommand{\gqlog}{\slog\gq}
\newcommand{\gqqlog}{\slog\gqq}
\newcommand{\gqqqlog}{\slog\gqqq}
\newcommand{\llog}{\slog<}
\newcommand{\ltlog}{\slog\lt}
\newcommand{\lttlog}{\slog\ltt}
\newcommand{\ltttlog}{\slog\lttt}
\newcommand{\glog}{\slog>}
\newcommand{\gtlog}{\slog\gt}
\newcommand{\gttlog}{\slog\gtt}
\newcommand{\gtttlog}{\slog\gttt}
%
% Miscellaneous
\newcommand{\LHS}{\mathrm{LHS}}
\newcommand{\RHS}{\mathrm{RHS}}
% .. operators
\DeclareMathOperator{\poly}{poly}
\DeclareMathOperator{\polylog}{polylog}
\DeclareMathOperator{\quasipoly}{quasipoly}
\DeclareMathOperator{\negl}{negl}
\DeclareMathOperator*{\argmin}{arg\,min}
\DeclareMathOperator*{\argmax}{arg\,max}
% .. functions
\DeclareMathOperator{\id}{id}
\DeclareMathOperator{\sign}{sign}
\DeclareMathOperator{\err}{err}
\DeclareMathOperator{\ReLU}{ReLU}
% .. analysis
\let\d\undefined
\newcommand{\d}{\operatorname{d}\mathopen{}}
\newcommand{\df}[2]{ {\f{\d #1}{\d #2}}}
\newcommand{\ds}[2]{ {\s{\d #1}{\d #2}}}
\newcommand{\part}{\partial}
\newcommand{\partf}[2]{\f{\part #1}{\part #2}}
\newcommand{\parts}[2]{\s{\part #1}{\part #2}}
\newcommand{\grad}[1]{\mathop{\nabla\!_{#1}}}
% .. sets of numbers
\newcommand{\N}{\mathbb{N}}
\newcommand{\Z}{\mathbb{Z}}
\newcommand{\R}{\mathbb{R}}
\newcommand{\C}{\mathbb{C}}
\newcommand{\F}{\mathbb{F}}
%
%%% SPECIALIZED MATH %%%
%
% Logic
\renewcommand{\and}{\wedge}
\newcommand{\AND}{\bigwedge}
\newcommand{\or}{\vee}
\newcommand{\OR}{\bigvee}
\newcommand{\xor}{\oplus}
\newcommand{\XOR}{\bigoplus}
\newcommand{\union}{\cup}
\newcommand{\inter}{\cap}
\newcommand{\UNION}{\bigcup}
\newcommand{\INTER}{\bigcap}
\newcommand{\comp}{\overline}
\newcommand{\true}{\r{true}}
\newcommand{\false}{\r{false}}
\newcommand{\tf}{\set{\true,\false}}
\DeclareMathOperator{\One}{\mathbb{1}}
\DeclareMathOperator{\1}{\mathbb{1}}
%
% Linear algebra
\renewcommand{\span}{\mathrm{span}}
\DeclareMathOperator{\rank}{rank}
\DeclareMathOperator{\proj}{proj}
\DeclareMathOperator{\dom}{dom}
\DeclareMathOperator{\Img}{Im}
\newcommand{\transp}{\mathsf{T}}
\renewcommand{\t}{^\transp}
% ... named tensors
\newcommand{\namedtensorstrut}{\vphantom{fg}} % milder than \mathstrut
\newcommand{\name}[1]{\mathsf{\namedtensorstrut #1}}
\newcommand{\nbin}[2]{\mathbin{\underset{\substack{#1}}{\namedtensorstrut #2}}}
\newcommand{\ndot}[1]{\nbin{#1}{\odot}}
\newcommand{\ncat}[1]{\nbin{#1}{\oplus}}
\newcommand{\nsum}[1]{\sum\limits_{\substack{#1}}}
\newcommand{\nfun}[2]{\mathop{\underset{\substack{#1}}{\namedtensorstrut\mathrm{#2}}}}
\newcommand{\ndef}[2]{\newcommand{#1}{\name{#2}}}
\newcommand{\nt}[1]{^{\transp(#1)}}
%
% Probability
\newcommand{\Normal}{\mathcal{N}}
\let\Pr\undefined
\DeclareMathOperator*{\Pr}{Pr}
\DeclareMathOperator*{\G}{\mathbb{G}}
\DeclareMathOperator*{\Odds}{Od}
\DeclareMathOperator*{\E}{E}
\DeclareMathOperator*{\Var}{Var}
\DeclareMathOperator*{\Cov}{Cov}
\DeclareMathOperator*{\corr}{corr}
\DeclareMathOperator*{\median}{median}
\newcommand{\dTV}{d_{\mathrm{TV}}}
\newcommand{\dHel}{d_{\mathrm{Hel}}}
\newcommand{\dJS}{d_{\mathrm{JS}}}
% ... information theory
\let\H\undefined
\DeclareMathOperator*{\H}{H}
\DeclareMathOperator*{\I}{I}
\DeclareMathOperator*{\D}{D}
%
%%% SPECIALIZED COMPUTER SCIENCE %%%
%
% Complexity classes
% .. classical
\newcommand{\Poly}{\mathsf{P}}
\newcommand{\NP}{\mathsf{NP}}
\newcommand{\PH}{\mathsf{PH}}
\newcommand{\PSPACE}{\mathsf{PSPACE}}
\renewcommand{\L}{\mathsf{L}}
% .. probabilistic
\newcommand{\formost}{\mathsf{Я}}
\newcommand{\RP}{\mathsf{RP}}
\newcommand{\BPP}{\mathsf{BPP}}
\newcommand{\MA}{\mathsf{MA}}
\newcommand{\AM}{\mathsf{AM}}
\newcommand{\IP}{\mathsf{IP}}
\newcommand{\RL}{\mathsf{RL}}
% .. circuits
\newcommand{\NC}{\mathsf{NC}}
\newcommand{\AC}{\mathsf{AC}}
\newcommand{\ACC}{\mathsf{ACC}}
\newcommand{\TC}{\mathsf{TC}}
\newcommand{\Ppoly}{\mathsf{P}/\poly}
\newcommand{\Lpoly}{\mathsf{L}/\poly}
% .. resources
\newcommand{\TIME}{\mathsf{TIME}}
\newcommand{\SPACE}{\mathsf{SPACE}}
\newcommand{\TISP}{\mathsf{TISP}}
\newcommand{\SIZE}{\mathsf{SIZE}}
% .. keywords
\newcommand{\co}{\mathsf{co}}
\newcommand{\Prom}{\mathsf{Promise}}
%
% Boolean analysis
\newcommand{\zo}{\set{0,1}}
\newcommand{\pmo}{\set{\pm 1}}
\newcommand{\zpmo}{\set{0,\pm 1}}
\newcommand{\harpoon}{\!\upharpoonright\!}
\newcommand{\rr}[2]{#1\harpoon_{#2}}
\newcommand{\Fou}[1]{\widehat{#1}}
\DeclareMathOperator{\Ind}{\mathrm{Ind}}
\DeclareMathOperator{\Inf}{\mathrm{Inf}}
\newcommand{\Der}[1]{\operatorname{D}_{#1}\mathopen{}}
\newcommand{\Exp}[1]{\operatorname{E}_{#1}\mathopen{}}
\DeclareMathOperator{\Stab}{\mathrm{Stab}}
\DeclareMathOperator{\T}{T}
\DeclareMathOperator{\sens}{\mathrm{s}}
\DeclareMathOperator{\bsens}{\mathrm{bs}}
\DeclareMathOperator{\fbsens}{\mathrm{fbs}}
\DeclareMathOperator{\Cert}{\mathrm{C}}
\DeclareMathOperator{\DT}{\mathrm{DT}}
\DeclareMathOperator{\CDT}{\mathrm{CDT}} % canonical
\DeclareMathOperator{\ECDT}{\mathrm{ECDT}}
\DeclareMathOperator{\CDTv}{\mathrm{CDT_{vars}}}
\DeclareMathOperator{\ECDTv}{\mathrm{ECDT_{vars}}}
\DeclareMathOperator{\CDTt}{\mathrm{CDT_{terms}}}
\DeclareMathOperator{\ECDTt}{\mathrm{ECDT_{terms}}}
\DeclareMathOperator{\CDTw}{\mathrm{CDT_{weighted}}}
\DeclareMathOperator{\ECDTw}{\mathrm{ECDT_{weighted}}}
\DeclareMathOperator{\AvgDT}{\mathrm{AvgDT}}
\DeclareMathOperator{\PDT}{\mathrm{PDT}} % partial decision tree
\DeclareMathOperator{\DTsize}{\mathrm{DT_{size}}}
\DeclareMathOperator{\W}{\mathbf{W}}
% .. functions (small caps sadly doesn't work)
\DeclareMathOperator{\Par}{\mathrm{Par}}
\DeclareMathOperator{\Maj}{\mathrm{Maj}}
\DeclareMathOperator{\HW}{\mathrm{HW}}
\DeclareMathOperator{\Thr}{\mathrm{Thr}}
\DeclareMathOperator{\Tribes}{\mathrm{Tribes}}
\DeclareMathOperator{\RotTribes}{\mathrm{RotTribes}}
\DeclareMathOperator{\CycleRun}{\mathrm{CycleRun}}
\DeclareMathOperator{\SAT}{\mathrm{SAT}}
\DeclareMathOperator{\UniqueSAT}{\mathrm{UniqueSAT}}
%
% Dynamic optimality
\newcommand{\OPT}{\mathsf{OPT}}
\newcommand{\Alt}{\mathsf{Alt}}
\newcommand{\Funnel}{\mathsf{Funnel}}
%
% Alignment
\DeclareMathOperator{\Amp}{\mathrm{Amp}}
%
%%% TYPESETTING %%%
%
% In text
\renewcommand{\th}{^{\mathrm{th}}}
\newcommand{\degree}{^\circ}
%
% Fonts
% .. bold
\newcommand{\BA}{\boldsymbol{A}}
\newcommand{\BB}{\boldsymbol{B}}
\newcommand{\BC}{\boldsymbol{C}}
\newcommand{\BD}{\boldsymbol{D}}
\newcommand{\BE}{\boldsymbol{E}}
\newcommand{\BF}{\boldsymbol{F}}
\newcommand{\BG}{\boldsymbol{G}}
\newcommand{\BH}{\boldsymbol{H}}
\newcommand{\BI}{\boldsymbol{I}}
\newcommand{\BJ}{\boldsymbol{J}}
\newcommand{\BK}{\boldsymbol{K}}
\newcommand{\BL}{\boldsymbol{L}}
\newcommand{\BM}{\boldsymbol{M}}
\newcommand{\BN}{\boldsymbol{N}}
\newcommand{\BO}{\boldsymbol{O}}
\newcommand{\BP}{\boldsymbol{P}}
\newcommand{\BQ}{\boldsymbol{Q}}
\newcommand{\BR}{\boldsymbol{R}}
\newcommand{\BS}{\boldsymbol{S}}
\newcommand{\BT}{\boldsymbol{T}}
\newcommand{\BU}{\boldsymbol{U}}
\newcommand{\BV}{\boldsymbol{V}}
\newcommand{\BW}{\boldsymbol{W}}
\newcommand{\BX}{\boldsymbol{X}}
\newcommand{\BY}{\boldsymbol{Y}}
\newcommand{\BZ}{\boldsymbol{Z}}
\newcommand{\Ba}{\boldsymbol{a}}
\newcommand{\Bb}{\boldsymbol{b}}
\newcommand{\Bc}{\boldsymbol{c}}
\newcommand{\Bd}{\boldsymbol{d}}
\newcommand{\Be}{\boldsymbol{e}}
\newcommand{\Bf}{\boldsymbol{f}}
\newcommand{\Bg}{\boldsymbol{g}}
\newcommand{\Bh}{\boldsymbol{h}}
\newcommand{\Bi}{\boldsymbol{i}}
\newcommand{\Bj}{\boldsymbol{j}}
\newcommand{\Bk}{\boldsymbol{k}}
\newcommand{\Bp}{\boldsymbol{p}}
\newcommand{\Bq}{\boldsymbol{q}}
\newcommand{\Br}{\boldsymbol{r}}
\newcommand{\Bs}{\boldsymbol{s}}
\newcommand{\Bt}{\boldsymbol{t}}
\newcommand{\Bu}{\boldsymbol{u}}
\newcommand{\Bv}{\boldsymbol{v}}
\newcommand{\Bw}{\boldsymbol{w}}
\newcommand{\Bx}{\boldsymbol{x}}
\newcommand{\By}{\boldsymbol{y}}
\newcommand{\Bz}{\boldsymbol{z}}
\newcommand{\Balpha}{\boldsymbol{\alpha}}
\newcommand{\Bbeta}{\boldsymbol{\beta}}
\newcommand{\Bgamma}{\boldsymbol{\gamma}}
\newcommand{\Bdelta}{\boldsymbol{\delta}}
\newcommand{\Beps}{\boldsymbol{\eps}}
\newcommand{\Bveps}{\boldsymbol{\veps}}
\newcommand{\Bzeta}{\boldsymbol{\zeta}}
\newcommand{\Beta}{\boldsymbol{\eta}}
\newcommand{\Btheta}{\boldsymbol{\theta}}
\newcommand{\Biota}{\boldsymbol{\iota}}
\newcommand{\Bkappa}{\boldsymbol{\kappa}}
\newcommand{\Blambda}{\boldsymbol{\lambda}}
\newcommand{\Bmu}{\boldsymbol{\mu}}
\newcommand{\Bnu}{\boldsymbol{\nu}}
\newcommand{\Bxi}{\boldsymbol{\xi}}
\newcommand{\Bomicron}{\boldsymbol{\omicron}}
\newcommand{\Bpi}{\boldsymbol{\pi}}
\newcommand{\Brho}{\boldsymbol{\rho}}
\newcommand{\Bsigma}{\boldsymbol{\sigma}}
\newcommand{\Btau}{\boldsymbol{\tau}}
\newcommand{\Bupsilon}{\boldsymbol{\upsilon}}
\newcommand{\Bphi}{\boldsymbol{\phi}}
\newcommand{\Bfi}{\boldsymbol{\fi}}
\newcommand{\Bchi}{\boldsymbol{\chi}}
\newcommand{\Bpsi}{\boldsymbol{\psi}}
\newcommand{\Bomega}{\boldsymbol{\omega}}
% .. calligraphic
\newcommand{\CA}{\mathcal{A}}
\newcommand{\CB}{\mathcal{B}}
\newcommand{\CC}{\mathcal{C}}
\newcommand{\CD}{\mathcal{D}}
\newcommand{\CE}{\mathcal{E}}
\newcommand{\CF}{\mathcal{F}}
\newcommand{\CG}{\mathcal{G}}
\newcommand{\CH}{\mathcal{H}}
\newcommand{\CI}{\mathcal{I}}
\newcommand{\CJ}{\mathcal{J}}
\newcommand{\CK}{\mathcal{K}}
\newcommand{\CL}{\mathcal{L}}
\newcommand{\CM}{\mathcal{M}}
\newcommand{\CN}{\mathcal{N}}
\newcommand{\CO}{\mathcal{O}}
\newcommand{\CP}{\mathcal{P}}
\newcommand{\CQ}{\mathcal{Q}}
\newcommand{\CR}{\mathcal{R}}
\newcommand{\CS}{\mathcal{S}}
\newcommand{\CT}{\mathcal{T}}
\newcommand{\CU}{\mathcal{U}}
\newcommand{\CV}{\mathcal{V}}
\newcommand{\CW}{\mathcal{W}}
\newcommand{\CX}{\mathcal{X}}
\newcommand{\CY}{\mathcal{Y}}
\newcommand{\CZ}{\mathcal{Z}}
% .. typewriter
\newcommand{\TA}{\mathtt{A}}
\newcommand{\TB}{\mathtt{B}}
\newcommand{\TC}{\mathtt{C}}
\newcommand{\TD}{\mathtt{D}}
\newcommand{\TE}{\mathtt{E}}
\newcommand{\TF}{\mathtt{F}}
\newcommand{\TG}{\mathtt{G}}
\newcommand{\TH}{\mathtt{H}}
\newcommand{\TI}{\mathtt{I}}
\newcommand{\TJ}{\mathtt{J}}
\newcommand{\TK}{\mathtt{K}}
\newcommand{\TL}{\mathtt{L}}
\newcommand{\TM}{\mathtt{M}}
\newcommand{\TN}{\mathtt{N}}
\newcommand{\TO}{\mathtt{O}}
\newcommand{\TP}{\mathtt{P}}
\newcommand{\TQ}{\mathtt{Q}}
\newcommand{\TR}{\mathtt{R}}
\newcommand{\TS}{\mathtt{S}}
\newcommand{\TT}{\mathtt{T}}
\newcommand{\TU}{\mathtt{U}}
\newcommand{\TV}{\mathtt{V}}
\newcommand{\TW}{\mathtt{W}}
\newcommand{\TX}{\mathtt{X}}
\newcommand{\TY}{\mathtt{Y}}
\newcommand{\TZ}{\mathtt{Z}}$
Personal thought process from reading (part of) Specifying Humans Formally (Using an Oracle for Physics) | Ordinary Ideas, Hazards for Formal Specifications | Ordinary Ideas, What does the universal prior actually look like? | Ordinary Ideas and The Solomonoff Prior is Malign — AI Alignment Forum.
Note: This hypothetical scenario is only relevant if we think we’ll at some point be able to faithfully approximate the process of finding the shortest Turing machine for a task, which is not only computationally intractable but actually uncomputable in general. Perhaps the most likely way we might be able do that is if we already have some fairly superintelligent machine that is very good at high-level mathematical reasoning? But it still seems overwhelmingly likely that thinking about this is a waste of time, so proceed at your own risk.
Suppose we want to replicate a human decision process in code by:
- Putting Alice in a box for a few years (sorry Alice!), give her some inputs $x$ through a computer interface, record its keyboard outputs $\newcommand{\ytype}{y_\text{typing}}\ytype$ and some brain patterns $\newcommand{\yscan}{y_\text{scan}}\yscan$ through an MRI scan.
- Select the shortest Turing machine $T$ that returns $y = (\ytype,\yscan)$ when run on $x$. This is basically equivalent to finding the best compression algorithm for $y$ given $x$ (but regularized by how complicated $T$ is).
- Now use $T$ as an emulation of Alice by giving it new inputs, and seeing what it outputs in $\ytype$ (either by re-runnning $T$, or just continuing to feed $T$ bits in case it works with a stream of input bits instead of a fixed input).
Maybe let’s think of $(x,y)$ as containing a few tens of gigabytes in total?
The hope is that the shortest $T$ will simulate Alice’s brain pretty faithfully (down to the level of detail that’s required to match her decisions and scan images) and straightforwardly. However, this is not the only solution. Let’s compare a few possible options for $T$ and see how much it needs to “encode”.
Solution 1.
- Description: Simulate Alice’s brain down to some level of detail, as well as the process of taking the MRI scan.
- Might need to encode:
- the general idea of this simulation and how it’s run
- how human brains generally work
- Alice’s memories, patterns of thoughts, goals, etc.
- this is what we care about recovering
- Alice’s anatomy
- how the MRI scanner works, and some quirks of how this particular scanner was manufactured
- inherent noise in the scan images
- possibly the largest part of the cost
Solution 2.
- Description: Simulate our entire universe from the start and describe which place in time and space we care about.
- Might need to encode:
- the laws of physics
- a very precise and (most likely) complete initial state of the universe
- this already seems much much larger than tens of gigabytes to me, but evidently Christiano disagrees with me in either the fact that this information is needed or that we need to know it very precisely in order to make the simulation work
- a description of the place and time at which the outputs are produced (i.e. the cables coming out of the box and when to look at them)
Solution 3.
- Description: Simulate some alternate universe $U$ where consequentialist intelligent life appears and then tries to hijack our experiments by strategically producing outputs in a canonical location of $U$.
- Might need to encode:
- the description of some universe sufficiently complex to evolve intelligent life
- (this should be very cheap)
- still most of the information in solution 1
- but crucially, we can save some bits of information while describing the setup because a consequentialist species would naturally think about
- what kind of experiments an intelligent species like us might a priori want to run
- what kinds of experiments we might want to run that are particularly interesting to hijack (e.g. because this allows them to make important decisions in our world or even break out of the simulation)
- what kinds of thought patterns a species who has evolved to be able to run these kinds of experiments might have (so they’d be able to have a good prior about the contents of Alice’s brains)
- some way to decide when to stop faithfully simulating Alice and start producing malign outputs
- this might be just deciding to be malign $99\%$ of the time, but hardcoding the randomness so that it falls within the $1\%$ for the specific $(x,y)$ that it’s selected for producing
It seems like solution 2 is way worse. On the other hand, solution 3 might end up being cheaper than solution 1 assuming that the cost of describing a universe that produces consequentialist life is lower than the savings we get from that consequentialist life guessing what kinds of experiments we might run (that are related to particularly important decisions) instead of writing these experiments out completely.
Current takes:
- somewhat convinced that this is plausible?
- but also it seems like some of the savings here could be attained by a much more stripped down / direct version of the cost-saving heuristics
- but on the other hand, maybe “just make a world that generates consequentialist life” is still the cheapest way to learn these heuristics
- and the particular bit about “what experiments are particularly important to hijack?” seems natural only if you’re a consequentialist agent
- other possible ways this could fail
- maybe most consequentialist life wouldn’t try to do this
- or might even choose to counterbalance the ill-intentioned ones by producing correct completions instead
- maybe consequentialist life is actually really hard to produce and relies on some pretty extraordinary coincidences?
- which would explain why we’re not seeing any aliens?
- my favorite explanation for not seeing aliens is just that self-replicating biological life is very rarely produced under our laws of physics
- and to the extent that we should believe anthropic arguments, our laws of physics are probably close to the ones that maximize the probability of spawning life at least as intelligent as ours
- which would mean that overall, we would need a lot of bits to describe a universe where intelligent consequentialist life pops up
- (and things like Conway’s Game of Life are fascinating but don’t seem to quite be in the same intuitive ballpark)
- how about we just use time-bounded Kolmogorov complexity instead?
- seems like it would solve the issue pretty conclusively
- and possibly much more tractable to approximate by a superintelligence that’s good at math
- general feeling: this seems very contingent on entropy quantities of very small numbers of bits and which are very hard to get any feeling about
- actually maybe this could be generalized to claim that basically any complicated enough input-output behavior that’s useful to us would involve simulating a universe that produces (consequential) intelligent life!
- if:
- it’s fairly easy to describe universes that produce intelligent life
- intelligent species end up caring in part about similar things/concepts there’s an intelligent-life data prior
- they all end up developing pretty good compression algorithms on their data (like we’re doing with LLMs)
- and it’s fairly easy to point to that compression algorithm within the simulation
- either because the life-form is consequential and places it in a canonical location explicitly to hijack the process
- or because intelligent life is just generally easy to spot and their compression algorithm is pretty important to them
- and the desired input-output behavior has significant mutual information with the intelligent-life data prior which is not easily explained by anything simpler
- then the shortest program describing that input-output behavior involves simulation of universes producing intelligent life