Results for "full pass through data"
Controlling robots via language.
Physical form contributes to computation.
Robots learning via exploration and growth.
Legal right to fair treatment.
AI giving legal advice without authorization.
Mechanics of price formation.
Returns above benchmark.
Predicting protein 3D structure from sequence.
Risk of incorrect financial models.
Rules governing auctions.
Agents fail to coordinate optimally.
AI tacitly coordinating prices.
Supplying buy/sell orders.
Groups adopting extreme positions.
Rate at which AI capabilities improve.
Tradeoff between safety and performance.
Intelligence and goals are independent.
Goals useful regardless of final objective.
Designing AI to cooperate with humans and each other.
A subfield of AI where models learn patterns from data to make predictions or decisions, improving with experience rather than explicit rule-coding.
A branch of ML using multi-layer neural networks to learn hierarchical representations, often excelling in vision, speech, and language.
Minimizing average loss on training data; can overfit when data is limited or biased.
Automatically learning useful internal features (latent variables) that capture salient structure for downstream tasks.
Separating data into training (fit), validation (tune), and test (final estimate) to avoid leakage and optimism bias.
Forcing predictable formats for downstream systems; reduces parsing errors and supports validation/guardrails.
Models that learn to generate samples resembling training data.
Learning from data generated by a different policy.
Model that compresses input into latent space and reconstructs it.
Trend reversal when data is aggregated improperly.
Shift in model outputs.