scalable methods
Techniques in AI that can handle increasing amounts of data and complexity without a proportional increase in computational resources or time, ensuring the viability of models and algorithms as they are deployed in real-world applications.
- Automaton Constrained Q-Learning
- Clip-and-Verify: Linear Constraint-Driven Domain Clipping for Accelerating Neural Network Verification
- Conditional Gradient Methods with Standard LMO for Stochastic Simple Bilevel Optimization
- DualCnst: Enhancing Zero-Shot Out-of-Distribution Detection via Text-Image Consistency in Vision-Language Models
- Enhancing Training Data Attribution with Representational Optimization
- Hankel Singular Value Regularization for Highly Compressible State Space Models
- STACI: Spatio-Temporal Aleatoric Conformal Inference