Weight Initialization

Backpropagation

ब्याकप्रोपागेसन

The algorithm that adjusts weights by sending error signals backward through the network.

Hyperparameter

हाइपरप्यारामिटर

A configuration value set before training that controls how a model learns, like learning rate.

Layer Normalization

लेयर नर्मलाइजेसन

Normalizing values across features within a layer to stabilize training.

Missing Data

हराएको डाटा

Values absent from a dataset that must be handled before modeling.

Parameter

प्यारामिटर

A value inside a model (like a weight) learned from data during training.

Pre-training

पूर्व-तालिम

The initial large-scale training of a model on broad data before any task-specific tuning.

Validation Set

प्रमाणीकरण सेट

Data used during training to tune settings and avoid overfitting before final testing.

Alignment

एलाइनमेन्ट / तालमेल

Ensuring an AI system's goals and behavior match human values and intentions.