Towards Faster Training Algorithms Exploiting Bandit Sampling From Convex to Strongly Convex Conditions.
Yangfan ZhouKaizhu HuangCheng ChengXuguang WangAmir HussainXin LiuPublished in: IEEE Trans. Emerg. Top. Comput. Intell. (2023)
Keyphrases
- convex optimization problems
- optimization problems
- piecewise linear
- random sampling
- efficient optimization
- sufficient conditions
- neural network
- convex optimization
- training set
- theoretical analysis
- computationally efficient
- significant improvement
- sampling algorithm
- convex functions
- computationally demanding
- machine learning
- data sets
- multi armed bandit