In this work, we introduce a modification to traditional reinforcement learning which we call observational dropout, whereby we limit the agents ability to observe the real environment at each timestep.

Large language models (LLMs) are increasingly trained with reinforcement learning from verifiable rewards (RLVR), yet real-world deployment demands models that can self-improve without labels or external judges. Existing self-improvement approaches primarily rely on self-confirmation signals (e.g., confidence, entropy, or consistency) to generate rewards. This reliance drives models toward over-confident, majority-favored solutions, causing an entropy collapse that degrades pass@n and reasoning complexity. To address this, we propose EVOL-RL, a label-free framework that mirrors the evolutionary principle of balancing selection with variation. Concretely, EVOL-RL retains the majority-voted answer as an anchor for stability, but adds a novelty-aware reward that scores each sampled solution by how different its reasoning is from other concurrently generated responses. This majority-for-stability + novelty-for-exploration rule mirrors the variation-selection principle: selection prevents drift, while novelty prevents collapse. Evaluation results show that EVOL-RL consistently outperforms the majority-only baseline; e.g., training on label-free AIME24 lifts Qwen3-4B-Base AIME25 pass@1 from baseline's 4.6% to 16.4%, and pass@16 from 18.5% to 37.9%. EVOL-RL not only prevents in-domain diversity collapse but also improves out-of-domain generalization (from math reasoning to broader tasks, e.g., GPQA, MMLU-Pro, and BBEH). The code is available at: https://github.com/YujunZhou/EVOL-RL.

Adversarial Imitation Learning (AIL) methods, while effective in settings with limited expert demonstrations, are often considered unstable. These approaches typically decompose into two components: Density Ratio (DR) estimation $\frac{ρ_E}{ρ_π}$, where a discriminator estimates the relative occupancy of state-action pairs under the policy versus the expert; and Reward Assignment (RA), where this ratio is transformed into a reward signal used to train the policy. While significant research has focused on improving density estimation, the role of reward assignment in influencing training dynamics and final policy performance has been largely overlooked. RA functions in AIL are typically derived from divergence minimization objectives, relying heavily on human design and ingenuity. In this work, we take a different approach: we investigate the discovery of data-driven RA functions, i.e, based directly on the performance of the resulting imitation policy. To this end, we leverage an LLM-guided evolutionary framework that efficiently explores the space of RA functions, yielding \emph{Discovered Adversarial Imitation Learning} (DAIL), the first meta-learnt AIL algorithm. Remarkably, DAIL generalises across unseen environments and policy optimization algorithms, outperforming the current state-of-the-art of \emph{human-designed} baselines. Finally, we analyse why DAIL leads to more stable training, offering novel insights into the role of RA functions in the stability of AIL. Code is publicly available: https://github.com/shshnkreddy/DAIL.

Understanding the relationships between objectives in a multiobjective optimisation problem is important for developing tailored and efficient solving techniques. In particular, when tackling combinatorial optimisation problems with many objectives, that arise in real-world logistic scenarios, better support for the decision maker can be achieved through better understanding of the often complex fitness landscape. This paper makes a contribution in this direction by presenting a technique that allows a visualisation and analysis of the local and global relationships between objectives in optimisation problems with many objectives. The proposed technique uses four steps: First, the global pairwise relationships are analysed using the Kendall correlation method; then, the ranges of the values found on the given Pareto front are estimated and assessed; next, these ranges are used to plot a map using Gray code, similar to Karnaugh maps, that has the ability to highlight the trade-offs between multiple objectives; and finally, local relationships are identified using scatter plots. Experiments are presented for three combinatorial optimisation problems: multiobjective multidimensional knapsack problem, multiobjective nurse scheduling problem, and multiobjective vehicle routing problem with time windows . Results show that the proposed technique helps in the gaining of insights into the problem difficulty arising from the relationships between objectives.

The diversity of patterns that emerge from complex systems motivates their use for scientific or artistic purposes. When exploring these systems, the challenges faced are the size of the parameter space and the strongly non-linear mapping between parameters and emerging patterns. In addition, artists and scientists who explore complex systems do so with an expectation of particular patterns. Taking these expectations into account adds a new set of challenges, which the exploration process must address. We provide design choices and their implementation to address these challenges; enabling the maximization of the diversity of patterns discovered in the user's region of interest -- which we call the constrained diversity -- in a sample-efficient manner. The region of interest is expressed in the form of explicit constraints. These constraints are formulated by the user in a system-agnostic way, and their addition enables interactive system exploration leading to constrained diversity, while maintaining global diversity.

Explainable AI (XAI) methods help identify which image regions influence a model's prediction, but often face a trade-off between detail and interpretability. Layer-wise Relevance Propagation (LRP) offers a model-aware alternative. However, LRP implementations commonly rely on heuristic rule sets that are not optimized for clarity or alignment with model behavior. We introduce EVO-LRP, a method that applies Covariance Matrix Adaptation Evolution Strategy (CMA-ES) to tune LRP hyperparameters based on quantitative interpretability metrics, such as faithfulness or sparseness. EVO-LRP outperforms traditional XAI approaches in both interpretability metric performance and visual coherence, with strong sensitivity to class-specific features. These findings demonstrate that attribution quality can be systematically improved through principled, task-specific optimization.

In certain real-world optimization scenarios, practitioners are not interested in solving multiple problems but rather in finding the best solution to a single, specific problem. When the computational budget is large relative to the cost of evaluating a candidate solution, multiple heuristic alternatives can be tried to solve the same given problem, each possibly with a different algorithm, parameter configuration, initialization, or stopping criterion. The sequential selection of which alternative to try next is crucial for efficiently identifying the one that provides the best possible solution across multiple attempts. Despite the relevance of this problem in practice, it has not yet been the exclusive focus of any existing review. Several sequential alternative selection strategies have been proposed in different research topics, but they have not been comprehensively and systematically unified under a common perspective. This work presents a focused review of single-problem multi-attempt heuristic optimization. It brings together suitable strategies to this problem that have been studied separately through algorithm selection, parameter tuning, multi-start and resource allocation. These strategies are explained using a unified terminology within a common framework, which supports the development of a taxonomy for systematically organizing and classifying them.

Humanoid robots have seen significant advancements in both design and control, with a growing emphasis on integrating these aspects to enhance overall performance. Traditionally, robot design has followed a sequential process, where control algorithms are developed after the hardware is finalized. However, this can be myopic and prevent robots to fully exploit their hardware capabilities. Recent approaches advocate for co-design, optimizing both design and control in parallel to maximize robotic capabilities. This paper presents the Evolutionary Continuous Adaptive RL-based Co-Design (EA-CoRL) framework, which combines reinforcement learning (RL) with evolutionary strategies to enable continuous adaptation of the control policy to the hardware. EA-CoRL comprises two key components: Design Evolution, which explores the hardware choices using an evolutionary algorithm to identify efficient configurations, and Policy Continuous Adaptation, which fine-tunes a task-specific control policy across evolving designs to maximize performance rewards. We evaluate EA-CoRL by co-designing the actuators (gear ratios) and control policy of the RH5 humanoid for a highly dynamic chin-up task, previously unfeasible due to actuator limitations. Comparative results against state-of-the-art RL-based co-design methods show that EA-CoRL achieves higher fitness score and broader design space exploration, highlighting the critical role of continuous policy adaptation in robot co-design.

Black-box optimization (BBO) addresses problems where objectives are accessible only through costly queries without gradients or explicit structure. Classical derivative-free methods -- line search, direct search, and model-based solvers such as Bayesian optimization -- form the backbone of BBO, yet often struggle in high-dimensional, noisy, or mixed-integer settings. Recent advances use machine learning (ML) and reinforcement learning (RL) to enhance BBO: ML provides expressive surrogates, adaptive updates, meta-learning portfolios, and generative models, while RL enables dynamic operator configuration, robustness, and meta-optimization across tasks. This paper surveys these developments, covering representative algorithms such as NNs with the modular model-based optimization framework (mlrMBO), zeroth-order adaptive momentum methods (ZO-AdaMM), automated BBO (ABBO), distributed block-wise optimization (DiBB), partition-based Bayesian optimization (SPBOpt), the transformer-based optimizer (B2Opt), diffusion-model-based BBO, surrogate-assisted RL for differential evolution (Surr-RLDE), robust BBO (RBO), coordinate-ascent model-based optimization with relative entropy (CAS-MORE), log-barrier stochastic gradient descent (LB-SGD), policy improvement with black-box (PIBB), and offline Q-learning with Mamba backbones (Q-Mamba). We also review benchmark efforts such as the NeurIPS 2020 BBO Challenge and the MetaBox framework. Overall, we highlight how ML and RL transform classical inexact solvers into more scalable, robust, and adaptive frameworks for real-world optimization.