DeCLIP (decoupled contrastive learning)
DeCLIP (decoupled contrastive learning) is an advanced approach in machine learning that enhances the efficiency of representation learning through decoupling tasks.
machine learning
Model merging (neural networks)
Model merging in neural networks is a technique that combines multiple trained neural network models into a single model. This approach aims to integrate knowledge from different models to improve performance, efficiency, or adaptability.
Waymo Open Dataset
The Waymo Open Dataset is a large-scale dataset designed for the development of autonomous vehicle technologies, featuring diverse driving scenarios and high-definition maps.
Logit bias
Logit bias refers to the intentional adjustment of the logit values in machine learning models to influence prediction probabilities. It is commonly used in natural language processing and classification tasks to control or steer model outputs.
AVA (atomic visual actions dataset)
The AVA dataset is a large-scale dataset for recognizing atomic visual actions in video clips, facilitating advancements in computer vision and machine learning.
Fisher-BRC (behavior regularized critic)
Fisher-BRC is a reinforcement learning framework that incorporates behavior regularization to enhance the learning process of an agent.
Concept activation vectors (CAV)
Concept activation vectors (CAV) are a method used in machine learning to interpret neural networks by associating specific directions in the latent space with human-understandable concepts.
Hyper-deep ensembles
Hyper-deep ensembles are advanced machine learning models that combine multiple deep neural networks to improve predictive performance, robustness, and uncertainty estimation. They extend traditional ensemble methods by leveraging very large or highly complex models in a coordinated manner.
SNLI (Stanford Natural Language Inference)
SNLI is a large-scale dataset designed to support research in natural language inference, containing labeled pairs of sentences.
Differentiable predictive coding (DPC)
Differentiable predictive coding (DPC) is a computational framework that integrates predictive coding principles with differentiable neural network architectures. It enables end-to-end learning of hierarchical prediction models by minimizing prediction errors through gradient-based optimization.