optimization process
The iterative procedure of enhancing a model's parameters to minimize or maximize an objective function, often involving techniques like gradient descent to reduce loss and improve performance, fundamentally underpinning the training of AI algorithms.
- FANS: A Flatness-Aware Network Structure for Generalization in Offline Reinforcement Learning
- Irrational Complex Rotations Empower Low-bit Optimizers
- Manipulating Feature Visualizations with Gradient Slingshots
- Optimization Inspired Few-Shot Adaptation for Large Language Models
- PROFIT: A Specialized Optimizer for Deep Fine Tuning
- Pattern-Guided Adaptive Prior for Structure Learning
- Safe RLHF-V: Safe Reinforcement Learning from Multi-modal Human Feedback
- Segment then Splat: Unified 3D Open-Vocabulary Segmentation via Gaussian Splatting
- Sparse MeZO: Less Parameters for Better Performance in Zeroth-Order LLM Fine-Tuning