Results for "distribution shift"
Distribution Shift
IntermediateTrain/test environment mismatch.
Distribution shift is like when you practice basketball in a gym but then have to play in a different setting, like outdoors on a windy day. The conditions have changed, and your skills might not work as well. In AI, this happens when a model is trained on one type of data but then faces differen...
Recovering training data from gradients.
Inferring sensitive features of training data.
Probabilistic energy-based neural network with hidden variables.
Probabilistic graphical model for structured prediction.
Diffusion performed in latent space for efficiency.
Simultaneous Localization and Mapping for robotics.
Monte Carlo method for state estimation.
Identifying abrupt changes in data generation.
Average value under a distribution.
Approximating expectations via random sampling.
Assigning a role or identity to the model.
Sampling multiple outputs and selecting consensus.
Prompt augmented with retrieved documents.
Randomizing simulation parameters to improve real-world transfer.
Enables external computation or lookup.
Maximum expected loss under normal conditions.
Fast approximation of costly simulations.
No agent can improve without hurting another.
Designing efficient marketplaces.
Inferring the agent’s internal state from noisy sensor data.