Difficulty: Intermediate

412 terms

Edge Inference Intermediate

Running models locally.
Embedding Intermediate

A continuous vector encoding of an item (word, image, user) such that semantic similarity corresponds to geometric closeness.
Emergent Abilities Intermediate

Capabilities that appear only beyond certain model sizes.
Emergent Coordination Intermediate

Coordination arising without explicit programming.
Empirical Risk Minimization Intermediate

Minimizing average loss on training data; can overfit when data is limited or biased.
Encryption in Transit/At Rest Intermediate

Protecting data during network transfer and while stored; essential for ML pipelines handling sensitive data.
Energy-Based Model Intermediate

Models that define an energy landscape rather than explicit probabilities.
Entropy Intermediate

A measure of randomness or uncertainty in a probability distribution.
Epoch Intermediate

One complete traversal of the training dataset during training.
EU AI Act Intermediate

European regulation classifying AI systems by risk.
Eval Harness Intermediate

System for running consistent evaluations across tasks, versions, prompts, and model settings.
Experiment Tracking Intermediate

Logging hyperparameters, code versions, data snapshots, and results to reproduce and compare experiments.
Explainability Intermediate

Techniques to understand model decisions (global or local), important in high-stakes and regulated settings.
Explainability Mandate Intermediate

Requirement to provide explanations.
Explainability Requirement Intermediate

Legal or policy requirement to explain AI decisions.
Explainable Credit Model Intermediate

Credit models with interpretable logic.
Exploding Gradient Intermediate

Gradients grow too large, causing divergence; mitigated by clipping, normalization, careful init.
Exploration-Exploitation Tradeoff Intermediate

Balancing learning new behaviors vs exploiting known rewards.
Exposure Bias Intermediate

Differences between training and inference conditions.
Expressivity Intermediate

The range of functions a model can represent.
F1 Score Intermediate

Harmonic mean of precision and recall; useful when balancing false positives/negatives matters.
Factor Graph Intermediate

Graphical model expressing factorization of a probability distribution.
Fair Lending Intermediate

Ensuring models comply with lending fairness laws.
False Negative Intermediate

Failure to detect present disease.
FDA Clearance Intermediate

US approval process for medical AI devices.
Feature Intermediate

A measurable property or attribute used as model input (raw or engineered), such as age, pixel intensity, or token ID.
Feature Engineering Intermediate

Designing input features to expose useful structure (e.g., ratios, lags, aggregations), often crucial outside deep learning.
Feature Store Intermediate

Centralized repository for curated features.
Federated Learning Intermediate

Training across many devices/silos without centralizing raw data; aggregates updates, not data.
Feedback Intermediate

Using output to adjust future inputs.