whpearson

Pos­si­ble Im­pli­ca­tions of the neu­ral retro­trans­posons to the future

Ac­ti­va­tion En­er­gies and Uncertainty

Crunch­course—a tool for com­bat­ing learn­ing akrasia

An In­tro­duc­tion to Con­trol Markets

Bayesian anal­y­sis un­der threat in Bri­tish courts

How good an em­u­la­tion is re­quired?

An­gles of Attack

Less­wrong UK plan­ning thread

Drive-less AIs and experimentation

De­bate: Is short term plan­ning in hu­mans due to a short life or due to bias?

Call for Ideas: In­dus­trial scale ex­is­ten­tial risk re­search

My prob­lems with For­mal Friendly Ar­tifi­cial In­tel­li­gence work

Pro­gram­matic Pre­dic­tion markets

Im­mod­esty, De­ceit and Mo­ti­vated Reasoning

Re­search Syndicates

Out­line of an ap­proach to AGI Estimation

Defin­ing the nor­mal com­puter con­trol problem

Mak­ing de­ci­sions in a real com­puter—an ar­gu­ment for meta-de­ci­sion the­ory research

Back of the en­velope calcu­la­tions around the sin­gu­lar­ity.

Re­quest for col­lab­o­ra­tors—Sur­vey on AI risk priorities