Results for "training loss"

AdvertisementAd space — search-top

188 results

Non-Convex Optimization Intermediate

Optimization with multiple local minima/saddle points; typical in neural networks.

AI Economics & Strategy
Second-Order Methods Intermediate

Optimization using curvature information; often expensive at scale.

AI Economics & Strategy
CLIP Intermediate

Joint vision-language model aligning images and text.

Computer Vision
Neural Vocoder Intermediate

Generates audio waveforms from spectrograms.

Speech & Audio AI
Feedback Loop Intermediate

Using production outcomes to improve models.

MLOps & Infrastructure
Gradient Advanced

Direction of steepest ascent of a function.

Mathematics
Line Search Intermediate

Choosing step size along gradient direction.

Foundations & Theory
Reflection Prompting Intro

Asking model to review and improve output.

Prompting & Instructions
Predictive Coding Frontier

Learning by minimizing prediction error.

World Models & Cognition
Safety-Critical System Advanced

Systems where failure causes physical harm.

Agents & Autonomy
Unauthorized Practice of Law Intermediate

AI giving legal advice without authorization.

AI in Law
Competitive Game Advanced

Agents have opposing objectives.

Agents & Autonomy
Fine-Tuning Intermediate

Updating a pretrained model’s weights on task-specific data to improve performance or adapt style/behavior.

Large Language Models
Differential Privacy Intermediate

A formal privacy framework ensuring outputs do not reveal much about any single individual’s data contribution.

Security & Privacy
Hyperparameters Intermediate

Configuration choices not learned directly (or not typically learned) that govern training or architecture.

Optimization
Batch Size Intermediate

Number of samples per gradient update; impacts compute efficiency, generalization, and stability.

Foundations & Theory
Warmup Intermediate

Gradually increasing learning rate at training start to avoid divergence.

AI Economics & Strategy
Gradient Leakage Intermediate

Recovering training data from gradients.

AI Economics & Strategy
Model Inversion Intermediate

Inferring sensitive features of training data.

AI Economics & Strategy
Exposure Bias Intermediate

Differences between training and inference conditions.

Model Failure Modes
Deceptive Alignment Advanced

Model behaves well during training but not deployment.

AI Safety & Alignment
Semi-Supervised Learning Intermediate

Training with a small labeled dataset plus a larger unlabeled dataset, leveraging assumptions like smoothness/cluster structure.

Machine Learning
Model Intermediate

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

Foundations & Theory
Underfitting Intermediate

When a model cannot capture underlying structure, performing poorly on both training and test data.

Foundations & Theory
Train/Validation/Test Split Intermediate

Separating data into training (fit), validation (tune), and test (final estimate) to avoid leakage and optimism bias.

Evaluation & Benchmarking
ReLU Intermediate

Activation max(0, x); improves gradient flow and training speed in deep nets.

Foundations & Theory
Exploding Gradient Intermediate

Gradients grow too large, causing divergence; mitigated by clipping, normalization, careful init.

Foundations & Theory
Weight Initialization Intermediate

Methods to set starting weights to preserve signal/gradient scales across layers.

Foundations & Theory
Dropout Intermediate

Randomly zeroing activations during training to reduce co-adaptation and overfitting.

Foundations & Theory
Data Augmentation Intermediate

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

Foundations & Theory

Welcome to AI Glossary

The free, self-building AI dictionary. Help us keep it free—click an ad once in a while!

Search

Type any question or keyword into the search bar at the top.

Browse

Tap a letter in the A–Z bar to browse terms alphabetically, or filter by domain, industry, or difficulty level.

3D WordGraph

Fly around the interactive 3D graph to explore how AI concepts connect. Click any word to read its full definition.