The unprecedented advancement of artificial intelligence has placed immense demands on computing hardware, but traditional silicon-based semiconductor technologies are approaching their physical and economic limit, prompting the exploration of novel computing paradigms. Memristor offers a promising solution, enabling in-memory analog computation and massive parallelism, which leads to low latency and power consumption. This manuscript reviews the current status of memristor-based machine learning accelerators, highlighting the milestones achieved in developing prototype chips, that not only accelerate neural networks inference but also tackle other machine learning tasks. More importantly, it discusses our opinion on current key challenges that remain in this field, such as device variation, the need for efficient peripheral circuitry, and systematic co-design and optimization. We also share our perspective on potential future directions, some of which address existing challenges while others explore untouched territories. By addressing these challenges through interdisciplinary efforts spanning device engineering, circuit design, and systems architecture, memristor-based accelerators could significantly advance the capabilities of AI hardware, particularly for edge applications where power efficiency is paramount.

This paper introduces HEPPO, an FPGA-based accelerator designed to optimize the Generalized Advantage Estimation (GAE) stage in Proximal Policy Optimization (PPO). Unlike previous approaches that focused on trajectory collection and actor-critic updates, HEPPO addresses GAE's computational demands with a parallel, pipelined architecture implemented on a single System-on-Chip (SoC). This design allows for the adaptation of various hardware accelerators tailored for different PPO phases. A key innovation is our strategic standardization technique, which combines dynamic reward standardization and block standardization for values, followed by 8-bit uniform quantization. This method stabilizes learning, enhances performance, and manages memory bottlenecks, achieving a 4x reduction in memory usage and a 1.5x increase in cumulative rewards. We propose a solution on a single SoC device with programmable logic and embedded processors, delivering throughput orders of magnitude higher than traditional CPU-GPU systems. Our single-chip solution minimizes communication latency and throughput bottlenecks, significantly boosting PPO training efficiency. Experimental results show a 30% increase in PPO speed and a substantial reduction in memory access time, underscoring HEPPO's potential for broad applicability in hardware-efficient reinforcement learning algorithms.

Differential evolution (DE) algorithm is recognized as one of the most effective evolutionary algorithms, demonstrating remarkable efficacy in black-box optimization due to its derivative-free nature. Numerous enhancements to the fundamental DE have been proposed, incorporating innovative mutation strategies and sophisticated parameter tuning techniques to improve performance. However, no single variant has proven universally superior across all problems. To address this challenge, we introduce a novel framework that employs reinforcement learning (RL) to automatically design DE for black-box optimization through meta-learning. RL acts as an advanced meta-optimizer, generating a customized DE configuration that includes an optimal initialization strategy, update rule, and hyperparameters tailored to a specific black-box optimization problem. This process is informed by a detailed analysis of the problem characteristics. In this proof-of-concept study, we utilize a double deep Q-network for implementation, considering a subset of 40 possible strategy combinations and parameter optimizations simultaneously. The framework's performance is evaluated against black-box optimization benchmarks and compared with state-of-the-art algorithms. The experimental results highlight the promising potential of our proposed framework.

This paper introduces an innovative drone carrier concept that is applied in maritime port security or offshore rescue. This system works with a heterogeneous system consisting of multiple Unmanned Aerial Vehicles (UAVs) and Unmanned Surface Vehicles (USVs) to perform inspection and intervention tasks in GNSS-denied or interrupted environments. The carrier, an electric catamaran measuring 4m by 7m, features a 4m by 6m deck supporting automated takeoff and landing for four DJI M300 drones, along with a 10kg-payload manipulator operable in up to level 3 sea conditions. Utilizing an offshore gimbal camera for navigation, the carrier can autonomously navigate, approach and dock with non-cooperative vessels, guided by an onboard camera, LiDAR, and Doppler Velocity Log (DVL) over a 3 km$^2$ area. UAVs equipped with onboard Ultra-Wideband (UWB) technology execute mapping, detection, and manipulation tasks using a versatile gripper designed for wet, saline conditions. Additionally, two UAVs can coordinate to transport large objects to the manipulator or interact directly with them. These procedures are fully automated and were successfully demonstrated at the Mohammed Bin Zayed International Robotic Competition (MBZIRC2024), where the drone carrier equipped with four UAVS and one manipulator, automatically accomplished the intervention tasks in sea-level-3 (wave height 1.25m) based on the rough target information.

Physics-Informed Neural Networks (PINNs) have revolutionized the computation of PDE solutions by integrating partial differential equations (PDEs) into the neural network's training process as soft constraints, becoming an important component of the scientific machine learning (SciML) ecosystem. In its current implementation, PINNs are mainly optimized using first-order methods like Adam, as well as quasi-Newton methods such as BFGS and its low-memory variant, L-BFGS. However, these optimizers often struggle with highly non-linear and non-convex loss landscapes, leading to challenges such as slow convergence, local minima entrapment, and (non)degenerate saddle points. In this study, we investigate the performance of Self-Scaled Broyden (SSBroyden) methods and other advanced quasi-Newton schemes, including BFGS and L-BFGS with different line search strategies approaches. These methods dynamically rescale updates based on historical gradient information, thus enhancing training efficiency and accuracy. We systematically compare these optimizers on key challenging linear, stiff, multi-scale and non-linear PDEs benchmarks, including the Burgers, Allen-Cahn, Kuramoto-Sivashinsky, and Ginzburg-Landau equations, and extend our study to Physics-Informed Kolmogorov-Arnold Networks (PIKANs) representation. Our findings provide insights into the effectiveness of second-order optimization strategies in improving the convergence and accurate generalization of PINNs for complex PDEs by orders of magnitude compared to the state-of-the-art.

Current time series forecasting methods can be broadly classified into two categories: Channel Independent (CI) and Channel Dependent (CD) strategies, both aiming to capture the complex dependencies within time series data. However, the CI strategy fails to exploit highly correlated covariate information, while the CD strategy integrates all dependencies, including irrelevant or noisy ones, thus compromising generalization. To mitigate these issues, recent works have introduced the Channel Clustering (CC) strategy by grouping channels with similar characteristics and applying different modeling techniques to each cluster. However, coarse-grained clustering cannot flexibly capture complex, time-varying interactions. Addressing the above challenges, we propose TimeFilter, a graph-based framework for adaptive and fine-grained dependency modeling. Specifically, after constructing the graph with the input sequence, TimeFilter filters out irrelevant correlations and preserves the most critical ones through patch-specific filtering. Extensive experiments on 13 real-world datasets from various application domains demonstrate the state-of-the-art performance of TimeFilter. The code is available at https://github.com/TROUBADOUR000/TimeFilter.

Accurate power forecasting from renewable energy sources (RES) is crucial for integrating additional RES capacity into the power system and realizing sustainability goals. This work emphasizes the importance of integrating decentralized spatio-temporal data into forecasting models. However, decentralized data ownership presents a critical obstacle to the success of such spatio-temporal models, and incentive mechanisms to foster data-sharing need to be considered. The main contributions are a) a comparative analysis of the forecasting models, advocating for efficient and interpretable spline LASSO regression models, and b) a bidding mechanism within the data/analytics market to ensure fair compensation for data providers and enable both buyers and sellers to express their data price requirements. Furthermore, an incentive mechanism for time series forecasting is proposed, effectively incorporating price constraints and preventing redundant feature allocation. Results show significant accuracy improvements and potential monetary gains for data sellers. For wind power data, an average root mean squared error improvement of over 10% was achieved by comparing forecasts generated by the proposal with locally generated ones.

This study proposes a novel approach for dynamic load balancing in Software-Defined Networks (SDNs) using a Transformer-based Deep Q-Network (DQN). Traditional load balancing mechanisms, such as Round Robin (RR) and Weighted Round Robin (WRR), are static and often struggle to adapt to fluctuating traffic conditions, leading to inefficiencies in network performance. In contrast, SDNs offer centralized control and flexibility, providing an ideal platform for implementing machine learning-driven optimization strategies. The core of this research combines a Temporal Fusion Transformer (TFT) for accurate traffic prediction with a DQN model to perform real-time dynamic load balancing. The TFT model predicts future traffic loads, which the DQN uses as input, allowing it to make intelligent routing decisions that optimize throughput, minimize latency, and reduce packet loss. The proposed model was tested against RR and WRR in simulated environments with varying data rates, and the results demonstrate significant improvements in network performance. For the 500MB data rate, the DQN model achieved an average throughput of 0.275 compared to 0.202 and 0.205 for RR and WRR, respectively. Additionally, the DQN recorded lower average latency and packet loss. In the 1000MB simulation, the DQN model outperformed the traditional methods in throughput, latency, and packet loss, reinforcing its effectiveness in managing network loads dynamically. This research presents an important step towards enhancing network performance through the integration of machine learning models within SDNs, potentially paving the way for more adaptive, intelligent network management systems.

Weaknesses: 1) The major weakness is that the paper can be seen as a simple extension of the Tensor Field Networks (TFN) [1] by adding attention layers, so novelty is somewhat limited. I do realize that some adaptations are necessary, e.g., using TFN layers to produce query/keys/values and showing that the attention weights are invariant, so there is value in the contributions. It's my understanding that the original TFN [1] was impractical for the point clouds sizes typically used for ModelNet and ShapeNet (a few thousand points). The proposed method includes a faster spherical harmonics implementation; is that part the bottleneck of the TFN? How big is the speed-up?

Unsupervised pre-training methods for large vision models have shown to enhance performance on downstream supervised tasks. Developing similar techniques for satellite imagery presents significant opportunities as unlabelled data is plentiful and the inherent temporal and multi-spectral structure provides avenues to further improve existing pre-training strategies. In this paper, we present SatMAE, a pre-training framework for temporal or multi-spectral satellite imagery based on Masked Autoencoder (MAE). To leverage temporal information, we include a temporal embedding along with independently masking image patches across time. In addition, we demonstrate that encoding multi-spectral data as groups of bands with distinct spectral positional encodings is beneficial.