optimization techniques
Optimization techniques refer to algorithms and methods used to find the best parameters for machine learning models, maximizing or minimizing a defined objective function, such as loss or error.
- An Optimized Franz-Parisi Criterion and its Equivalence with SQ Lower Bounds
- Differentiable Constraint-Based Causal Discovery
- Enabling Differentially Private Federated Learning for Speech Recognition: Benchmarks, Adaptive Optimizers, and Gradient Clipping
- Enforcing convex constraints in Graph Neural Networks
- Fast Last-Iterate Convergence of SGD in the Smooth Interpolation Regime
- Fine-Tuning Discrete Diffusion Models with Policy Gradient Methods
- Generalization Bounds for Model-based Algorithm Configuration
- Global Convergence for Average Reward Constrained MDPs with Primal-Dual Actor Critic Algorithm
- HermesFlow: Seamlessly Closing the Gap in Multimodal Understanding and Generation
- Improving Decision Trees through the Lens of Parameterized Local Search
- Inexact Column Generation for Bayesian Network Structure Learning via Difference-of-Submodular Optimization
- MOOSE-Chem2: Exploring LLM Limits in Fine-Grained Scientific Hypothesis Discovery via Hierarchical Search
- Mind the Gap: Bridging Thought Leap for Improved Chain-of-Thought Tuning
- Mixed-Sample SGD: an End-to-end Analysis of Supervised Transfer Learning
- Not All Data are Good Labels: On the Self-supervised Labeling for Time Series Forecasting
- Optimistic Online-to-Batch Conversions for Accelerated Convergence and Universality
- PhySense: Sensor Placement Optimization for Accurate Physics Sensing
- ProfiX: Improving Profile-Guided Optimization in Compilers with Graph Neural Networks
- QiMeng-NeuComBack: Self-Evolving Translation from IR to Assembly Code
- VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifold
- Which Algorithms Have Tight Generalization Bounds?