Constrained optimization of the parameters of a simulator plays a crucial role in a design process. These problems become challenging when the simulator is stochastic, computationally expensive, and the parameter space is high-dimensional. One can efficiently perform optimization only by utilizing the gradient with respect to the parameters, but these gradients are unavailable in many legacy, black-box codes. We introduce the algorithm Scout-Nd (Stochastic Constrained Optimization for N dimensions) to tackle the issues mentioned earlier by efficiently estimating the gradient, reducing the noise of the gradient estimator, and applying multi-fidelity schemes to further reduce computational effort. We validate our approach on standard benchmarks, demonstrating its effectiveness in optimizing parameters highlighting better performance compared to existing methods.

The convergence of deterministic policy gradient under the Hadamard parameterization is studied in the tabular setting and the linear convergence of the algorithm is established. To this end, we first show that the error decreases at an $O(\frac{1}{k})$ rate for all the iterations. Based on this result, we further show that the algorithm has a faster local linear convergence rate after $k_0$ iterations, where $k_0$ is a constant that only depends on the MDP problem and the initialization. To show the local linear convergence of the algorithm, we have indeed established the contraction of the sub-optimal probability $b_s^k$ (i.e., the probability of the output policy $\pi^k$ on non-optimal actions) when $k\ge k_0$.

In consequential domains, it is often impossible to compel individuals to take treatment, so that optimal policy rules are merely suggestions in the presence of human non-adherence to treatment recommendations. In these same domains, there may be heterogeneity both in who responds in taking-up treatment, and heterogeneity in treatment efficacy. While optimal treatment rules can maximize causal outcomes across the population, access parity constraints or other fairness considerations can be relevant in the case of encouragement. For example, in social services, a persistent puzzle is the gap in take-up of beneficial services among those who may benefit from them the most. When in addition the decision-maker has distributional preferences over both access and average outcomes, the optimal decision rule changes. We study causal identification, statistical variance-reduced estimation, and robust estimation of optimal treatment rules, including under potential violations of positivity. We consider fairness constraints such as demographic parity in treatment take-up, and other constraints, via constrained optimization. Our framework can be extended to handle algorithmic recommendations under an often-reasonable covariate-conditional exclusion restriction, using our robustness checks for lack of positivity in the recommendation. We develop a two-stage algorithm for solving over parametrized policy classes under general constraints to obtain variance-sensitive regret bounds. We illustrate the methods in two case studies based on data from randomized encouragement to enroll in insurance and from pretrial supervised release with electronic monitoring.

The rise of renewables coincides with the shift towards Electrical Vehicles (EVs) posing technical and operational challenges for the energy balance of the local grid. Nowadays, the energy grid cannot deal with a spike in EVs usage leading to a need for more coordinated and grid aware EVs charging and discharging strategies. However, coordinating power flow from multiple EVs into the grid requires sophisticated algorithms and load-balancing strategies as the complexity increases with more control variables and EVs, necessitating large optimization and decision search spaces. In this paper, we propose an EVs fleet coordination model for the day ahead aiming to ensure a reliable energy supply and maintain a stable local grid, by utilizing EVs to store surplus energy and discharge it during periods of energy deficit. The optimization problem is addressed using Harris Hawks Optimization (HHO) considering criteria related to energy grid balancing, time usage preference, and the location of EV drivers. The EVs schedules, associated with the position of individuals from the population, are adjusted through exploration and exploitation operations, and their technical and operational feasibility is ensured, while the rabbit individual is updated with a non-dominated EV schedule selected per iteration using a roulette wheel algorithm. The solution is evaluated within the framework of an e-mobility service in Terni city. The results indicate that coordinated charging and discharging of EVs not only meet balancing service requirements but also align with user preferences with minimal deviations. The assessment of the determined solutions' quality and efficacy shows promising outcomes, with convergence after 100 iterations reflected in a generational distance of 0.35 and a Pareto front error of 1.01, while the distribution of solutions exhibits strong hypervolume thus covering a significant portion of the objective space.

Micro scanning mirrors (MSM) extend the range and field of view of LiDARs, medical imaging devices, and laser projectors. However, a new class of soft-hinged MSMs contains out-of-plane translation in addition to the 2 degree-of-freedom rotations, which presents a cabliration challenge. We report a new calibration system and algorithm design to address the challenge. In the calibration system, a new low-cost calibration rig design employs a minimal 2-laser beam approach. The new new algorithm builds on the reflection principle and an optimization approach to precisely measure MSM poses. To establish the mapping between Hall sensor readings and MSM poses, we propose a self-synchronizing periodicity-based model fitting calibration approach. We achieve an MSM poses estimation accuracy of 0.020{\deg} with a standard deviation of 0.011{\deg}.

We study the theoretical aspects of Reinforced Language Models (RLMs) from a bi-objective optimization perspective. Specifically, we consider the RLMs as a Pareto optimization problem that maximizes the two conflicting objectives, i.e., reward objective and likelihood objectives, simultaneously. Our main contribution consists of three parts. First, we establish the theoretical foundations of RLM as a Pareto optimization problem by presenting Reward Upper BOund (RUBO) and Pareto optimality. Our theoretical outcomes are supported by not only deductive proofs but also empirical results. Second, we propose Reward Dropout, a simple yet powerful method that guarantees to improve a bi-objective optimization of RLM. Lastly, we demonstrate that the Reward Dropout is consistently effective across five benchmark datasets and four benchmark LLMs, meaning that the Reward Dropout significantly improves the optimization performance of RLMs.

Regarding state estimation, many works have explored The combination of cameras and inertial measurement units to use multiple vision sensors for better VINS performance (IMUs) have become prevalent in autonomous vehicles and (Leutenegger et al. 2015; Usenko et al. 2016; Paul mobile devices in the recent decade due to their decrease in et al. 2017; Sun et al. 2018; Kuo et al. 2020; Campos cost and complementary sensing nature. A camera provides et al. 2021; Fu et al. 2021). In particular, Leutenegger texture-rich images of 2 degree-of-freedom (DoF) bearing et al. (2015), Usenko et al. (2016) and Fu et al. (2021) observations to environmental features, while a 6-axis IMU have shown that stereo camera or multiple cameras can typically consists of a gyroscope and an accelerometer achieve better pose accuracy or lower the uncertainties which measures high-frequency angular velocity and linear of IMU-Camera calibration. Only a few works recently acceleration, respectively. This has lead to a significant investigate multiple inertial sensor fusion for VINS (Kim progress of developing visual-inertial navigation system et al. 2017; Eckenhoff et al. 2019b; Zhang et al. 2020; (VINS) algorithms focusing on efficient and accurate pose Wu et al. 2023; Faizullin and Ferrer 2023), showing that estimation (Huang 2019). While many works have shown the system robustness and pose accuracy can be improved accurate estimation for the minimal sensing case of a single by fusing additional IMUs. For optimal fusion of multiple camera and IMU (Mourikis and Roumeliotis 2007; Bloesch asynchronous visual and inertial sensors for MVIS, et al. 2015; Forster et al. 2016; Qin et al. 2018; Geneva et al. it is crucial to provide accurate full-parameter calibration 2020), it is known that the inclusion of additional sensors for these sensors, which include: (i) IMU-IMU/camera can provide improved accuracy due to additional information rigid transformation, (ii) IMU-IMU/camera time offset, (iii) and robustness to single sensor failure cases (Paul et al.

Pre-training has achieved remarkable success when transferred to downstream tasks. In machine learning, we care about not only the good performance of a model but also its behavior under reasonable shifts of condition. The same philosophy holds when pre-training a foundation model. However, the foundation model may not uniformly behave well for a series of related downstream tasks. This happens, for example, when conducting mask recovery regression where the recovery ability or the training instances diverge like pattern features are extracted dominantly on pre-training, but semantic features are also required on a downstream task. This paper considers pre-training a model that guarantees a uniformly good performance over the downstream tasks. We call this goal as $\textit{downstream-task robustness}$. Our method first separates the upstream task into several representative ones and applies a simple minimax loss for pre-training. We then design an efficient algorithm to solve the minimax loss and prove its convergence in the convex setting. In the experiments, we show both on large-scale natural language processing and computer vision datasets our method increases the metrics on worse-case downstream tasks. Additionally, some theoretical explanations for why our loss is beneficial are provided. Specifically, we show fewer samples are inherently required for the most challenging downstream task in some cases.

Recent vision-language models have achieved tremendous progress far beyond what we ever expected. However, their computational costs are also dramatically growing with rapid development, especially for the large models. It makes model acceleration exceedingly critical in a scenario of limited resources. Although extensively studied for unimodal models, the acceleration for multimodal models, especially the vision-language Transformers, is relatively under-explored. To pursue more efficient and accessible vision-language Transformers, this paper introduces \textbf{Cross}-\textbf{G}uided \textbf{E}nsemble of \textbf{T}okens (\textbf{\emph{CrossGET}}), a universal acceleration framework for vision-language Transformers. This framework adaptively combines tokens through real-time, cross-modal guidance, thereby achieving substantial acceleration while keeping high performance. \textit{CrossGET} has two key innovations: 1) \textit{Cross-Guided Matching and Ensemble}. \textit{CrossGET} incorporates cross-modal guided token matching and ensemble to exploit cross-modal information effectively, only introducing cross-modal tokens with negligible extra parameters. 2) \textit{Complete-Graph Soft Matching}. In contrast to the existing bipartite soft matching approach, \textit{CrossGET} introduces a complete-graph soft matching policy to achieve more reliable token-matching results while maintaining parallelizability and high efficiency. Extensive experiments are conducted on various vision-language tasks, including image-text retrieval, visual reasoning, image captioning, and visual question answering. Performance on both classic multimodal architectures and emerging multimodal LLMs demonstrate the effectiveness and versatility of the proposed \textit{CrossGET} framework. The code will be at \url{https://github.com/sdc17/CrossGET}.

In-hand object reorientation is necessary for performing many dexterous manipulation tasks, such as tool use in less structured environments that remain beyond the reach of current robots. Prior works built reorientation systems assuming one or many of the following: reorienting only specific objects with simple shapes, limited range of reorientation, slow or quasistatic manipulation, simulation-only results, the need for specialized and costly sensor suites, and other constraints which make the system infeasible for real-world deployment. We present a general object reorientation controller that does not make these assumptions. It uses readings from a single commodity depth camera to dynamically reorient complex and new object shapes by any rotation in real-time, with the median reorientation time being close to seven seconds. The controller is trained using reinforcement learning in simulation and evaluated in the real world on new object shapes not used for training, including the most challenging scenario of reorienting objects held in the air by a downward-facing hand that must counteract gravity during reorientation. Our hardware platform only uses open-source components that cost less than five thousand dollars. Although we demonstrate the ability to overcome assumptions in prior work, there is ample scope for improving absolute performance. For instance, the challenging duck-shaped object not used for training was dropped in 56 percent of the trials. When it was not dropped, our controller reoriented the object within 0.4 radians (23 degrees) 75 percent of the time. Videos are available at: https://taochenshh.github.io/projects/visual-dexterity.