AAAI 2026 accepted papers on Optimization
25 of the 4,417 papers accepted at AAAI 2026 (0.6%) are on Optimization, compared with 1.2% at AAAI 2025. They fall into 4 topics.
See AAAI 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Nonconvex and smooth optimization · 12 papers
- A Natural-Gradient Approach for Nonlinear Stochastic Systems with Parameter Uncertainty
- AT-Field: Rethinking the Games in Adversarial Training
- Convergence of Fast Policy Iteration in Markov Games and Robust MDPs
- Decentralized Non-convex Stochastic Optimization with Heterogeneous Variance
- ECPv2: Fast, Efficient, and Scalable Global Optimization of Lipschitz Functions
- Exploring Non-Convex Discrete Energy Landscapes: An Efficient Langevin-Like Sampler with Replica Exchange
- High-Order Error Bounds for Markovian LSA with Richardson–Romberg Extrapolation
- Methods for Optimization Problems with Markovian Stochasticity and Non-Euclidean Geometry
- Parallelizable Riemannian Alternating Direction Method of Multipliers for Non-convex Pose Graph Optimization
- Parameter-free Optimal Rates for Nonlinear Semi-Norm Contractions with Applications to Q-Learning
- Stochastic Decentralized Optimization of Non-Smooth Convex and Convex-Concave Problems over Time-Varying Networks
- Gaussian Approximation for Two-Timescale Linear Stochastic Approximation
Implicit bias and generalization · 5 papers
- Beyond Sharpness: The Role of Nonuniformity in Generalization
- Neural Tangent Kernels Under Stochastic Data Augmentation
- On the Learning Dynamics of Two-layer Linear Networks with Label Noise SGD
- Tractable Sharpness-Aware Learning of Probabilistic Circuits
- Unpacking the Implicit Norm Dynamics of Sharpness-Aware Minimization in Tensorized Models
Optimizers · 4 papers
Distributed and compressed · 4 papers
- DeToNATION: Decoupled Torch Network-Aware Training on Interlinked Online Nodes
- High Dimensional Distributed Gradient Descent with Arbitrary Number of Byzantine Attackers
- Ordered Local Momentum for Asynchronous Distributed Learning Under Arbitrary Delays
- S-D-RSM: Stochastic Distributed Regularized Splitting Method for Large-Scale Convex Optimization Problems