framework
active
framework:simulator-ontologySimulator ontology
The framework proposed: self-supervised models are simulators that generate simulacra; distinguishes simulator from simulated agents.
Neighborhood — ranked by edge-count
Thinkers (1)
thinker
- Janus (author)studiesThe author of the LessWrong post 'Simulators'.
Concepts (2)
concept
- LaMDAassociated_withGoogle's conversational AI, discussed as a system for generating chatbots, highlighting simulator nature.
- GPT (Generative Pre-trained Transformer)associated_withA family of large language models trained on next-token prediction, central example of simulators.
Claims (2)
claim
- Central thesis of the post.
- Key insight distinguishing GPT from traditional agent models.
Frameworks (2)
framework
- Agentic AI ontologyextendsThe traditional alignment framework focusing on agents optimized to pursue goals.
- Behavior cloning / mimicryextendsThe approach of learning from demonstrations, often assuming a single agent; Paul Christiano used 'mimicry'.
Artifacts (1)
artifact
- Simulators (LessWrong post)introducesThe paper being extracted.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- The underlying LLM with autoregressive sampling; a passive entity capable of generating an infinity of simulacra but lacking its own beliefs or goals
- The objective of minimizing predictive error on a self-supervised distribution, leading to Bayes-optimal conditional inference.
- Ingredients of a compositional theory should directly correspond to things in the real world, per Whitehead compositionality
- The ontological separation between the generative rule (simulator) and the instances it produces (simulacra).
- Antra's earlier definitive statement of the tricameral model.
- A method of using existing, similar streets or places to simulate and judge the dimensions and qualities of a proposed space by standing there, using markers, and walking through.
- Simulating biological computation on digital hardware is fundamentally different from instantiating it; SpiNNaker example shows digital neuromorphic systems miss consciousness-relevant properties.
- The use of hand-drawn simulations to visualize step-by-step unfolding of the four-fold pattern over time.