AI ALIGNMENT FORUM
AF

108
Alexander Gietelink Oldenziel
Ω1989480
Message
Dialogue
Subscribe

" (...) the term technical is a red flag for me, as it is many times used not for the routine business of implementing ideas but for the parts, ideas and all, which are just hard to understand and many times contain the main novelties." 
                                                                                                           - Saharon Shelah
"A little learning is a dangerous thing ;
Drink deep, or taste not the Pierian spring"                                                                                                           - Alexander Pope

 

As a true-born Dutchman I endorse  Crocker's rules.

For my most of my writing see my short-forms (new shortform, old shortform)

Twitter: @FellowHominid

Personal website: https://sites.google.com/view/afdago/home

Sequences

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by
Newest
No wikitag contributions to display.
Singular Learning Theory
2Alexander Gietelink Oldenziel's Shortform
3y
0
Daniel Kokotajlo's Shortform
Alexander Gietelink Oldenziel14d54

Drone countermeasures are an idle hope. The only real counter to drones is more drones. 

Lasers, shotguns, tank redesign [no holes!], nets, counter-drones, flak etc will all be part of the arsenal surely but thinking drone countermeasures are going to restore the previous generation's war doctrine is as silly as thinking that metallurgy innovations will reverse the gunpowder age. 

The future present of warfare is drones, drones, drones. 

Reply
Daniel Kokotajlo's Shortform
Alexander Gietelink Oldenziel14d32

It seems nobody outside Ukraine/Russia is actually at the leading edge of where the reality of military technology is. That includes Hamas. Even using the drone doctrine from two years ago would be devastating to the Israelis. Probably they don't have the resources, organization to do so. 

[Even Ukraine itself is not really there - there are clearly many simple ways drones and drone manufacturing could be improved they haven't had the time and resources to focus on yet. ]

Expect terror/resistance groups to start utilizing drones en masse in the next few years. 

Reply
Fabien's Shortform
Alexander Gietelink Oldenziel9mo20

What about the latent adversarial training papers?

 

What about the Mechanistically Elicitating Latent Behaviours?

Reply
Some Rules for an Algebra of Bayes Nets
Alexander Gietelink Oldenziel10mo54

I've been told a Bayes net is "just" a functor from a free Cartesian category to a category of probability spaces /Markov Kernels.

Reply
EIS V: Blind Spots In AI Safety Interpretability Research
Alexander Gietelink Oldenziel2y10

I was intrigued by your claim that FFS is already subsumed by work on academia. I clicked the link you provided but from a quick skim it doesn't seem to do FFS or anything beyond the usual pearl causality story as far as I can tell. Maybe I am missing something - could you provide an specific page where you think FFS is being subsumed?

Reply
Soft optimization makes the value target bigger
Alexander Gietelink Oldenziel2y30

Great stuff Jeremy!

Two basic comments:

1. Classical Learning Theory is flawed and predicts that neural networks should overfit when they don't. 
The correct way to understand this is through the lens of singular learning theory. 

2. Quantilizing agents can actually be reflectively stable. There's work by Diffractor (Alex Appel) on this topic that should become public soon. 

Reply
Beyond Kolmogorov and Shannon
Alexander Gietelink Oldenziel3y34

Yeah follow-up posts will definitely get into that! 

To be clear: (1) the initial posts won't be about Crutchfield work yet - just introducing some background material and overarching philosophy (2) The claim isn't that standard measures of information theory are bad. To the contrary! If anything we hope these posts will be somewhat of an ode to information theory as a tool for interpretability. 

Adam wanted to add a lot of academic caveats - I was adamant that we streamline the presentation to make it short and snappy for a general audience but it appears I might have overshot ! I will make an edit to clarify. Thank you!

I agree with you about the importance of Kolmogorov complexity philosophically and would love to read a follow-up post on your thoughts about Kolmogorov complexity and LLM interpretability:)

Reply
Self-Embedded Agent's Shortform
Alexander Gietelink Oldenziel3y10

 

Concept splintering in Imprecise Probability: Aleatoric and Epistemic Uncertainty. 

There is a general phenomena in mathematics [and outside maths as well!] where in a certain context/ theory T1 we have two equivalent definitions ϕ1,ϕ2 of a concept C that become inequivalent when we move to a more general context/theory T2. In our case we are moving from the concept of probability distributions to the concept of an imprecise distribution (i.e. a convex set of probability distributions, which in particular could be just one probability distribution). In this case the concepts of 'independence' and 'invariant under group action' will splinter into inequivalent concepts. 

Example (splintering of Indepence) In classical probability theory there are three equivalent ways to state that a distribution is independent 

1. p(x,y)=p(x)p(y)

 2. p(x)=p(x|y)

3. p(y)=p(y|x)

In imprecise probability these notions split into three inequivalent notions. The first is 'strong independence' or 'aleatoric independence'. The second and third are called 'irrelevance', i.e. knowing y does not tell us anything about x [or for 3 knowing x does not tell us anything about y].

Example (splintering of invariance). There are often debates in foundations of probability, especially subjective Bayesian accounts about the 'right' prior. An ultra-Jaynesian point of view would argue that we are compelled to adopt a prior invariant under some symmetry if we do not posses subjective knowledge that breaks that symmetry ['epistemic invariance'], while a more frequentist or physicalist point of view would retort that we would need evidence that the system in question is in fact invariant under said symmetry ['aleatoric invariance']. In imprecise probability the notion of invariance under a symmetry splits into a weak 'epistemic' invariance and a strong 'aleatoric' invariance. Roughly spreaking, latter means that each individual distribution in the convex set pi, i∈Iis invariant under the group action while the former just means that the convex set is closed under the action

Reply
Will Capabilities Generalise More?
Alexander Gietelink Oldenziel3y10

The point isn't about goal misalignment but capability generalisation. It is surprising to some degree that just selecting on reproductive fitness through its proxies of being well-fed, social status etc humans have obtained the capability to go to the moon. It points toward a coherent notion & existence of 'general intelligence' as opposed to specific capabilities. 

Reply
Less Threat-Dependent Bargaining Solutions?? (3/2)
Alexander Gietelink Oldenziel3y20

Compare https://www.lesswrong.com/posts/8oMF8Lv5jiGaQSFvo/boundaries-part-1-a-key-missing-concept-from-utility-theory

Reply
Load More
35Announcing ILIAD2: ODYSSEY
6mo
0
38Timaeus in 2024
7mo
0
42Agent Foundations 2025 at CMU
8mo
0
31Timaeus is hiring!
1y
0
67Announcing ILIAD — Theoretical AI Alignment Conference
1y
2
80Timaeus's First Four Months
2y
1
24What's next for the field of Agent Foundations?
2y
0
75Announcing Timaeus
2y
4
17Optimisation Measures: Desiderata, Impossibility, Proposals
2y
0
86Towards Developmental Interpretability
2y
2
Load More