Los Modelos Basados en Energía (EBMs) son un tipo de modelo probabilístico utilizado en aprendizaje automático and inteligencia artificial to represent complex distributions over data. They operate on the principle that each configuration of the model corresponds to an energy value, with lower energy indicating a more likely or favorable configuration. The key idea is to learn an energy function that assigns a scalar value to each possible data point, which can then be used to derive probabilities through normalization.
En términos matemáticos, un EBM define un probability distribution by associating an energy value, denoted as E(x), to each data point x. The probability of a particular data point is calculated using the Boltzmann distribution, which is expressed as:
P(x) = exp(-E(x)) / Z
Aquí, Z es la constante de normalización conocida como la función de partición, which ensures that the probabilities sum to one across all configurations. Learning in EBMs typically involves optimizing the energy function, often using techniques like contrastive divergence or other sampling methods.
Se ha demostrado que los EBMs son efectivos en varias aplicaciones, incluyendo generación de imágenes, denoising, and as generative models for aprendizaje no supervisado. They can capture complex relationships in the data, making them a powerful tool in the field of deep learning and beyond.