N NexChron

Training

Definition The process of adjusting a model's parameters by exposing it to data and minimizing a loss function using gradient-based optimization. Training is the computationally intensive phase that produces a model capable of making useful predictions.

In Depth

Large model pre-training can consume millions of GPU-hours and tens of millions of dollars. The training loop iterates over batches of data, computes a loss (e.g., cross-entropy for next-token prediction), back-propagates gradients, and updates weights via optimizers like AdamW. Training decisions — learning rate schedules, batch size, data mixtures — profoundly affect the quality of the resulting model.

Browse more terms

AI Agent AI Alignment Agentic AI Artificial General Intelligence Artificial Superintelligence Attention Mechanism AutoML Benchmark Bias (AI) Chain-of-Thought Computer Vision Deepfake Diffusion Model Edge AI Embedding Federated Learning Few-Shot Learning Fine-Tuning Function Calling GPT