Statements (22)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:person
|
| gptkbp:education |
gptkb:Harvard_University
gptkb:Moscow_Institute_of_Physics_and_Technology |
| gptkbp:employer |
gptkb:DeepMind
|
| gptkbp:field |
gptkb:artificial_intelligence
AI safety |
| gptkbp:founder |
gptkb:Future_of_Life_Institute
|
| gptkbp:hasBlog |
https://vkrakovna.wordpress.com/
|
| gptkbp:hasWebsite |
https://vkrakovna.wordpress.com/
|
| gptkbp:knownFor |
AI safety research
reward modeling specification gaming |
| gptkbp:nationality |
gptkb:Canadian
gptkb:Russian |
| gptkbp:occupation |
gptkb:researchers
gptkb:AI_safety_researcher |
| gptkbp:publishedIn |
Specification gaming examples in AI
Penalizing side effects using stepwise relative reachability |
| gptkbp:twitter |
@vkrakovna
|
| gptkbp:bfsParent |
gptkb:Future_of_Life_Institute
|
| gptkbp:bfsLayer |
6
|
| https://www.w3.org/2000/01/rdf-schema#label |
Viktoriya Krakovna
|