jessicata(Jessica Taylor)

Karma: 8,733

Jessica Taylor. CS undergrad and Master’s at Stanford; former research fellow at MIRI.

I work on decision theory, social epistemology, strategy, naturalized agency, mathematical foundations, decentralized networking systems and applications, theory of mind, and functional programming languages.

Blog: unstableontology.com

Twitter: https://twitter.com/jessi_cata

Probabilistic Oracle Machines and Nash Equilibria

jessicata6 Feb 2015 1:14 UTC

5 points

0 comments1 min readLW link

UDT in the Land of Probabilistic Oracles

jessicata8 Feb 2015 9:13 UTC

4 points

1 comment3 min readLW link

Statistical learning theory and robust concept learning

jessicata7 Mar 2015 4:13 UTC

4 points

6 comments4 min readLW link

Forum Digest: Reflective Oracles

jessicata22 Mar 2015 4:02 UTC

6 points

0 comments3 min readLW link

Reflective oracles and the procrastination paradox

jessicata26 Mar 2015 22:18 UTC

3 points

4 comments2 min readLW link

Minimax as an approach to reduced-impact AI

jessicata2 Apr 2015 22:00 UTC

5 points

3 comments4 min readLW link

Learning a concept using only positive examples

jessicata28 Apr 2015 3:57 UTC

8 points

6 comments4 min readLW link

Reflective probabilistic logic cannot assign positive probability to its own coherence and an inner reflection principle

jessicata7 May 2015 21:00 UTC

7 points

5 comments3 min readLW link

MIRIx Stanford report

jessicata11 May 2015 6:11 UTC

2 points

2 comments3 min readLW link

PA+100 cannot always predict modal UDT

jessicata12 May 2015 20:26 UTC

3 points

3 comments3 min readLW link

A problem with resource-bounded Solomonoff induction and unpredictable environments

jessicata27 Jul 2015 3:03 UTC

3 points

5 comments3 min readLW link

Quantilizers maximize expected utility subject to a conservative cost constraint

jessicata28 Sep 2015 2:17 UTC

33 points

3 comments5 min readLW link

Conservative classifiers

jessicata2 Oct 2015 3:56 UTC

3 points

0 comments6 min readLW link

A first look at the hard problem of corrigibility

jessicata15 Oct 2015 20:16 UTC

11 points

5 comments4 min readLW link

What do we need value learning for?

jessicata29 Nov 2015 1:41 UTC

3 points

0 comments3 min readLW link

Three preference frameworks for goal-directed agents

jessicata2 Dec 2015 0:06 UTC

4 points

1 comment3 min readLW link

A sketch of a value-learning sovereign

jessicata20 Dec 2015 21:32 UTC

12 points

15 comments13 min readLW link

Another view of quantilizers: avoiding Goodhart’s Law

jessicata9 Jan 2016 4:02 UTC

26 points

2 comments2 min readLW link

A possible training procedure for human-imitators

jessicata16 Feb 2016 22:43 UTC

4 points

4 comments3 min readLW link

A scheme for safely handling a mixture of good and bad predictors

jessicata17 Feb 2016 5:35 UTC

0 points

3 comments2 min readLW link