Results for "learning like humans"

Optimization under uncertainty.

Model trained on its own outputs degrades quality.

RL using learned or known environment models.

Predicts next state given current state and action.

Learning action mapping directly from demonstrations.

Identifying suspicious transactions.

AI applied to scientific problems.

Awareness and regulation of internal processes.

A mismatch between training and deployment data distributions that can degrade model performance.

The relationship between inputs and outputs changes over time, requiring monitoring and model updates.

A parameterized mapping from inputs to outputs; includes architecture + learned parameters.

A gradient method using random minibatches for efficient training on large datasets.

Popular optimizer combining momentum and per-parameter adaptive step sizes via first/second moment estimates.

Gradients shrink through layers, slowing learning in early layers; mitigated by ReLU, residuals, normalization.

Techniques that stabilize and speed training by normalizing activations; LayerNorm is common in Transformers.

A high-priority instruction layer setting overarching behavior constraints for a chat model.

A preference-based training method optimizing policies directly from pairwise comparisons without explicit RL loops.

Model trained to predict human preferences (or utility) for candidate outputs; used in RLHF-style pipelines.

Systematic differences in model outcomes across groups; arises from data, labels, and deployment context.

Feature attribution method grounded in cooperative game theory for explaining predictions in tabular settings.

Measure of consistency across labelers; low agreement indicates ambiguous tasks or poor guidelines.

When some classes are rare, requiring reweighting, resampling, or specialized metrics.

Expanding training data via transformations (flips, noise, paraphrases) to improve robustness.

Policies and practices for approving, monitoring, auditing, and documenting models in production.

Standardized documentation describing intended use, performance, limitations, data, and ethical considerations.

Structured dataset documentation covering collection, composition, recommended uses, biases, and maintenance.

Techniques that fine-tune small additional components rather than all weights to reduce compute and storage.

Observing model inputs/outputs, latency, cost, and quality over time to catch regressions and drift.

PEFT method injecting trainable low-rank matrices into layers, enabling efficient fine-tuning.

Inputs crafted to cause model errors or unsafe behavior, often imperceptible in vision or subtle in text.