Time series forecasting is extensively applied across diverse domains. Transformer-based models demonstrate significant potential in modeling cross-time and cross-variable interaction. However, we notice that the cross-variable correlation of multivariate time series demonstrates multifaceted (positive and negative correlations) and dynamic progression over time, which is not well captured by existing Transformer-based models. To address this issue, we propose a TimeCNN model to refine cross-variable interactions to enhance time series forecasting. Its key innovation is timepoint-independent, where each time point has an independent convolution kernel, allowing each time point to have its independent model to capture relationships among variables. This approach effectively handles both positive and negative correlations and adapts to the evolving nature of variable relationships over time. Extensive experiments conducted on 12 real-world datasets demonstrate that TimeCNN consistently outperforms state-of-the-art models. Notably, our model achieves significant reductions in computational requirements (approximately 60.46%) and parameter count (about 57.50%), while delivering inference speeds 3 to 4 times faster than the benchmark iTransformer model

Neural Information Processing Systems http://nips.cc/

Accurate calibration of gravity adaptation zones is of great significance in fields such as underwater navigation, geophysical exploration, and marine engineering. With the increasing application of gravity field data in these areas, traditional calibration methods based on single features are becoming inadequate for capturing the complex characteristics of gravity fields and addressing the intricate interrelationships among multidimensional data. This paper proposes an attention-enhanced algorithm for gravity adaptation zone calibration. By introducing an attention mechanism, the algorithm adaptively fuses multidimensional gravity field features and dynamically assigns feature weights, effectively solving the problems of multicollinearity and redundancy inherent in traditional feature selection methods, significantly improving calibration accuracy and robustness.In addition, a large-scale gravity field dataset with over 10,000 sampling points was constructed, and Kriging interpolation was used to enhance the spatial resolution of the data, providing a reliable data foundation for model training and evaluation. We conducted both qualitative and quantitative experiments on several classical machine learning models (such as SVM, GBDT, and RF), and the results demonstrate that the proposed algorithm significantly improves performance across these models, outperforming other traditional feature selection methods. The method proposed in this paper provides a new solution for gravity adaptation zone calibration, showing strong generalization ability and potential for application in complex environments. The code is available at \href{this link} {https://github.com/hulnifox/RF-ATTN}.

A BSTRACT In sparse reward scenarios of reinforcement learning (RL), the memory mechanism provides promising shortcuts to policy optimization by reflecting on past experiences like humans. However, current memory-based RL methods simply store and reuse high-value policies, lacking a deeper refining and filtering of diverse past experiences and hence limiting the capability of memory. In this paper, we propose AdaMemento, an adaptive memory-enhanced RL framework. Instead of just memorizing positive past experiences, we design a memory-reflection module that exploits both positive and negative experiences by learning to predict known local optimal policies based on real-time states. To effectively gather informative trajectories for the memory, we further introduce a fine-grained intrinsic motivation paradigm, where nuances in similar states can be precisely distinguished to guide exploration. The exploitation of past experiences and exploration of new policies are then adaptively coordinated by ensemble learning to approach the global optimum. Furthermore, we theoretically prove the superiority of our new intrinsic motivation and ensemble mechanism. From 59 quantitative and visualization experiments, we confirm that AdaMemento can distinguish subtle states for better exploration and effectively exploiting past experiences in memory, achieving significant improvement over previous methods. However, in sparse reward environments, policy updates become unstable and ineffective due to insufficient feedback (Bellemare et al., 2016; Liang et al., 2018). This significantly increases the difficulty of learning effective long-horizon policies. Memory offers a promising solution to the sparse reward problem, as humans can effectively learn from past experiences to avoid repeating mistakes in similar scenarios (Liu et al., 2021; Bransford & Johnson, 1972; Andrychowicz et al., 2017). Through memory, agents can utilize prior successful experiences to refine their policies in complex environments, hence reducing the reliance on dense reward feedback and improving both learning efficiency and policy stability (Pathak et al., 2017). Existing memory-based RL methods can be roughly categorized into two classes.

Model-based reinforcement learning has shown promise for improving sample efficiency and decision-making in complex environments. However, existing methods face challenges in training stability, robustness to noise, and computational efficiency. In this paper, we propose Bisimulation Metric for Model Predictive Control (BS-MPC), a novel approach that incorporates bisimulation metric loss in its objective function to directly optimize the encoder. This time-step-wise direct optimization enables the learned encoder to extract intrinsic information from the original state space while discarding irrelevant details and preventing the gradients and errors from diverging. BS-MPC improves training stability, robustness against input noise, and computational efficiency by reducing training time. We evaluate BS-MPC on both continuous control and image-based tasks from the DeepMind Control Suite, demonstrating superior performance and robustness compared to state-of-the-art baseline methods.

Path-parametric methods have gained popularity in the formulation of navigation algorithms, such as high-level planners [1, 2, 3], reinforcement learning (RL) policies [4, 5, 6] or low-level model predictive controllers (MPC) [7, 8, 9]. The fundamental concept behind these parametric methods is to either introduce the path parameter as an additional degree of freedom, enabling the system to regulate its progress along the path [10, 11], or to conduct a change of coordinates that project the Euclidean states to the spatial states, i.e., the progress along the path and the orthogonal distance to it [12, 13, 14]. These parametric formulations have proven successful for three primary reasons: firstly, they inherently capture the notion of advancement along the path, secondly they allow for embedding the path's geometric features, such as the curvature and the torsion, into the system dynamics, and thirdly, spatial bounds manifest as convex constraints in the orthogonal terms of the spatial states. Given the broad range of problems encompassing path-parametric approaches, existing methods remain detached from each other and are frequently presented as independent work. This has resulted in a disjointed body of literature, where these techniques are viewed as distinct methods. Consequently, the reader is left with a fragmented view of the path-parametric problem, making it difficult to understand the interplay between the different techniques. To close this gap, in this paper we show how all these approaches are interconnected by presenting a universal formulation for path-parametric planning and control. To this end, the path-parametric problem is analyzed from three different yet interconnected perspectives (i-iii): First, in Section 2, we study the (i) interplay of existing parametric techniques and show how they can be unified under a single framework consisting of two ingredients: (ii) a path-parameterization technique and (iii) a spatial representation of the system dynamics. These are discussed in-depth across the subsequent Sections 3 and 4, respectively.

The problem of uncertainty is a feature of real world robotics problems and any control framework must contend with it in order to succeed in real applications tasks. Reinforcement Learning is no different, and epistemic uncertainty arising from model uncertainty or misspecification is a challenge well captured by the sim-to-real gap. A simple solution to this issue is domain randomization (DR), which unfortunately can result in conservative agents. As a remedy to this conservativeness, the use of universal policies that take additional information about the randomized domain has risen as an alternative solution, along with recurrent neural network-based controllers. Uncertainty-aware universal policies present a particularly compelling solution able to account for system identification uncertainties during deployment. In this paper, we reveal that the challenge of efficiently optimizing uncertainty-aware policies can be fundamentally reframed as solving the convex coverage set (CCS) problem within a multi-objective reinforcement learning (MORL) context. By introducing a novel Markov decision process (MDP) framework where each domain's performance is treated as an independent objective, we unify the training of uncertainty-aware policies with MORL approaches. This connection enables the application of MORL algorithms for domain randomization (DR), allowing for more efficient policy optimization. To illustrate this, we focus on the linear utility function, which aligns with the expectation in DR formulations, and propose a series of algorithms adapted from the MORL literature to solve the CCS, demonstrating their ability to enhance the performance of uncertainty-aware policies.

Machine learning is an important part of the data science field. In petrophysics, machine learning algorithms and applications have been widely approached. In this context, Vietnam Petroleum Institute (VPI) has researched and deployed several effective prediction models, namely missing log prediction, fracture zone and fracture density forecast, etc. As one of our solutions, VPI-MLogs is a web-based deployment platform which integrates data preprocessing, exploratory data analysis, visualisation and model execution. Using the most popular data analysis programming language, Python, this approach gives users a powerful tool to deal with the petrophysical logs section. The solution helps to narrow the gap between common knowledge and petrophysics insights. This article will focus on the web-based application which integrates many solutions to grasp petrophysical data.

Controlling greenhouse crop production systems is a complex task due to uncertain and non-linear dynamics between crops, indoor and outdoor climate, and economics. The declining number of skilled growers necessitates the development of autonomous greenhouse control systems. Reinforcement Learning (RL) is a promising approach that can learn a control policy to automate greenhouse management. RL optimises a control policy through interactions with a model of the greenhouse while guided by an economic-based reward function. However, its application to real-world systems is limited due to discrepancies between models and real-world dynamics. Moreover, RL controllers may struggle to maintain state constraints while optimising the primary objective, especially when models inadequately capture the adverse effects of constraint violations on crop growth. Also, the generalisation to novel states, for example, due to unseen weather trajectories, is underexplored in RL-based greenhouse control. This work addresses these challenges through three key contributions. First, we present GreenLight-Gym, the first open-source environment designed for training and evaluating RL algorithms on the state-of-the-art greenhouse model GreenLight. GreenLight-Gym enables the community to benchmark RL-based control methodologies. Second, we compare two reward-shaping approaches, using either a multiplicative or additive penalty, to enforce state boundaries. The additive penalty achieves more stable training while better adhering to state constraints, while the multiplicative penalty yields marginally higher profits. Finally, we evaluate RL performance on a disjoint training and testing weather dataset, demonstrating improved generalisation to unseen conditions. Our environment and experiment scripts are open-sourced, facilitating innovative research on learning-based greenhouse control.

This contribution reports on a software framework that uses physically-based rendering to simulate camera operation in lunar conditions. The focus is on generating synthetic images qualitatively similar to those produced by an actual camera operating on a vehicle traversing and/or actively interacting with lunar terrain, e.g., for construction operations. The highlights of this simulator are its ability to capture (i) light transport in lunar conditions and (ii) artifacts related to the vehicle-terrain interaction, which might include dust formation and transport. The simulation infrastructure is built within an in-house developed physics engine called Chrono, which simulates the dynamics of the deformable terrain-vehicle interaction, as well as fallout of this interaction. The Chrono::Sensor camera model draws on ray tracing and Hapke Photometric Functions. We analyze the performance of the simulator using two virtual experiments featuring digital twins of NASA's VIPER rover navigating a lunar environment, and of the NASA's RASSOR excavator engaged into a digging operation. The sensor simulation solution presented can be used for the design and testing of perception algorithms, or as a component of in-silico experiments that pertain to large lunar operations, e.g., traversability, construction tasks.