TFPnP: Tuning-free Plug-and-Play Proximal Algorithms with Applications to Inverse Imaging Problems.
Kaixuan WeiAngelica I. Avilés-RiveroJingwei LiangYing FuHua HuangCarola-Bibiane SchönliebPublished in: J. Mach. Learn. Res. (2022)
Keyphrases
- optimization problems
- benchmark problems
- test problems
- computational problems
- tunable parameters
- combinatorial optimization
- problems in computer vision
- np hard problems
- theoretical analysis
- run times
- computer vision
- hard problems
- theoretical guarantees
- iterative algorithms
- approximate solutions
- difficult problems
- practical problems
- related problems
- parameter tuning
- search methods
- data sets
- learning algorithm
- evolutionary algorithm
- orders of magnitude
- benchmark datasets
- machine learning algorithms
- np complete
- computational cost
- significant improvement
- high resolution
- computational geometry
- specific problems
- image analysis
- data structure
- bayesian networks
- image processing
- graph theory