The field of optimization and decision-making is moving towards developing more efficient and adaptive methods for real-world applications. Recent research has focused on improving the trade-off between computational efficiency and statistical optimality, leading to the development of jointly efficient algorithms that can achieve nearly optimal regret bounds with low computational costs. Another area of advancement is in cost-aware stopping rules for Bayesian optimization, which can adapt to varying evaluation costs and provide theoretical guarantees on the expected cumulative evaluation cost. Additionally, there is a growing interest in developing variance-dependent bounds for regression problems, which can lead to more accurate predictions and better decision-making. Noteworthy papers include:

• A paper proposing a jointly efficient algorithm for generalized linear bandits that achieves a nearly optimal regret bound with low computational costs.
• A paper introducing a cost-aware stopping rule for Bayesian optimization that adapts to varying evaluation costs and provides theoretical guarantees on the expected cumulative evaluation cost.
• A paper presenting a novel loss function called the betting loss that achieves a variance-dependent bound for [0,1]-valued regression problems.