Search

Reinforcement learning utilizing kernel ridge regression to predict the expected value function represents a powerful method with great representational capacity. This setting is a highly versatile framework amenable to analytical results. We consider kernel-based function approximation for RL in the infinite horizon average reward setting, also referred to as the undiscounted setting. We propose an optimistic algorithm, similar to acquisition function based algorithms in the special case of bandits. We establish novel no-regret performance guarantees for our algorithm, under kernel-based modelling assumptions. Additionally, we derive a novel confidence interval for the kernel-based prediction of the expected value function, applicable across various RL problems., Comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)

Multi-Layer Perceptrons (MLP) are powerful tools for representing complex, non-linear relationships, making them essential for diverse machine learning and AI applications. Efficient hardware implementation of MLPs can be achieved through many hardware and architectural design techniques. These networks excel at predictive modeling and classification tasks like image classification, making them a popular choice. Approximate computing techniques are increasingly used to optimize critical path delay, area, power, and overall hardware efficiency in high-performance computing systems through controlled error and related trade-offs. This study proposes a hardware MLP neural network implemented in 45nm CMOS technology, in which MAC units of the neurons incorporate error and power controllable approximate multipliers for classification of the MNIST dataset. The optimized network consists of 10 neurons within the hidden layers, occupying 0.026mm2 of area, with 5.55mW at 100MHz frequency in accurate mode and 4.81mW in lowest accuracy mode. The experiments indicate that the proposed design achieves a maximum rate of 13.33% decrease overall and 24.78% in each neuron's power consumption with only a 0.92% decrease in accuracy in comparison with accurate circuit., Comment: 6 pages, 7 figures, accepted at The 6th Iranian International Conference on Microelectronics (IICM2024)

We investigate the effect of a Heaviside cut-off on the front propagation dynamics of the so-called Burgers-FisherKolmogoroff-Petrowskii-Piscounov (Burgers-FKPP) advection-reaction-diffusion equation. We prove the existence and uniqueness of a travelling front solution in the presence of a cut-off in the reaction kinetics and the advection term, and we derive the leading-order asymptotics for the speed of propagation of the front in dependence on the advection strength and the cut-off parameter. Our analysis relies on geometric techniques from dynamical systems theory and specifically, on geometric desingularisation, which also known as blow-up.

The limited energy available in most embedded systems poses a significant challenge in enhancing the performance of embedded processors and microcontrollers. One promising approach to address this challenge is the use of approximate computing, which can be implemented in both hardware and software layers to balance the trade-off between performance and power consumption. In this study, the impact of dynamic hardware approximation methods on the run-time energy efficiency of a RISC-V embedded processor with specialized features for approximate computing is investigated. The results indicate that the platform achieves an average energy efficiency of 13.3 pJ/instruction at a 500MHz clock frequency adhering approximation in 45nm CMOS technology. Compared to accurate circuits and computation, the approximate computing techniques in the processing core resulted in a significant improvement of 9.21% in overall energy efficiency, 60.83% in multiplication instructions, 14.64% in execution stage, and 9.23% in overall power consumption., Comment: 6 pages, 11 figures, accepted at The 6th Iranian International Conference on Microelectronics (IICM2024)

The demand for energy-efficient and high performance embedded systems drives the evolution of new hardware architectures, including concepts like approximate computing. This paper presents a novel reconfigurable embedded platform named "phoeniX", using the standard RISC-V ISA, maximizing energy efficiency while maintaining acceptable application-level accuracy. The platform enables the integration of approximate circuits at the core level with diverse structures, accuracies, and timings without requiring modifications to the core, particularly in the control logic. The platform introduces novel control features, allowing configurable trade-offs between accuracy and energy consumption based on specific application requirements. To evaluate the effectiveness of the platform, experiments were conducted on a set of applications, such as image processing and Dhrystone benchmark. The core with its original execution engine, occupies 0.024mm2 of area, with average power consumption of 4.23mW at 1.1V operating voltage, average energy-efficiency of 7.85pJ per operation at 620MHz frequency in 45nm CMOS technology. The configurable platform with a highly optimized 3-stage pipelined RV32I(E)M architecture, possesses a DMIPS/MHz of 1.89, and a CPI of 1.13, showcasing remarkable capabilities for an embedded processor., Comment: 9 Pages, 7 figures, 2024 27th Euromicro Conference on Digital System Design (DSD), Paris, France, 2024

We consider the problem of learning a realization of a partially observed dynamical system with linear state transitions and bilinear observations. Under very mild assumptions on the process and measurement noises, we provide a finite time analysis for learning the unknown dynamics matrices (up to a similarity transform). Our analysis involves a regression problem with heavy-tailed and dependent data. Moreover, each row of our design matrix contains a Kronecker product of current input with a history of inputs, making it difficult to guarantee persistence of excitation. We overcome these challenges, first providing a data-dependent high probability error bound for arbitrary but fixed inputs. Then, we derive a data-independent error bound for inputs chosen according to a simple random design. Our main results provide an upper bound on the statistical error rates and sample complexity of learning the unknown dynamics matrices from a single finite trajectory of bilinear observations., Comment: 35 pages, 3 figures

In the face of surging power demands for exascale HPC systems, this work tackles the critical challenge of understanding the impact of software-driven power management techniques like Dynamic Voltage and Frequency Scaling (DVFS) and Power Capping. These techniques have been actively developed over the past few decades. By combining insights from GPU benchmarking to understand application power profiles, we present a telemetry data-driven approach for deriving energy savings projections. This approach has been demonstrably applied to the Frontier supercomputer at scale. Our findings based on three months of telemetry data indicate that, for certain resource-constrained jobs, significant energy savings (up to 8.5%) can be achieved without compromising performance. This translates to a substantial cost reduction, equivalent to 1438 MWh of energy saved. The key contribution of this work lies in the methodology for establishing an upper limit for these best-case scenarios and its successful application. This work sheds light on potential energy savings and empowers HPC professionals to optimize the power-performance trade-off within constrained power budgets, not only for the exascale era but also beyond.

Semantic segmentation, as a crucial component of complex visual interpretation, plays a fundamental role in autonomous vehicle vision systems. Recent studies have significantly improved the accuracy of semantic segmentation by exploiting complementary information and developing multimodal methods. Despite the gains in accuracy, multimodal semantic segmentation methods suffer from high computational complexity and low inference speed. Therefore, it is a challenging task to implement multimodal methods in driving applications. To address this problem, we propose the Cosine Similarity Fusion Network (CSFNet) as a real-time RGB-X semantic segmentation model. Specifically, we design a Cosine Similarity Attention Fusion Module (CS-AFM) that effectively rectifies and fuses features of two modalities. The CS-AFM module leverages cross-modal similarity to achieve high generalization ability. By enhancing the fusion of cross-modal features at lower levels, CS-AFM paves the way for the use of a single-branch network at higher levels. Therefore, we use dual and single-branch architectures in an encoder, along with an efficient context module and a lightweight decoder for fast and accurate predictions. To verify the effectiveness of CSFNet, we use the Cityscapes, MFNet, and ZJU datasets for the RGB-D/T/P semantic segmentation. According to the results, CSFNet has competitive accuracy with state-of-the-art methods while being state-of-the-art in terms of speed among multimodal semantic segmentation models. It also achieves high efficiency due to its low parameter count and computational complexity. The source code for CSFNet will be available at https://github.com/Danial-Qashqai/CSFNet.

Reinforcement Learning (RL) has shown great empirical success in various application domains. The theoretical aspects of the problem have been extensively studied over past decades, particularly under tabular and linear Markov Decision Process structures. Recently, non-linear function approximation using kernel-based prediction has gained traction. This approach is particularly interesting as it naturally extends the linear structure, and helps explain the behavior of neural-network-based models at their infinite width limit. The analytical results however do not adequately address the performance guarantees for this case. We will highlight this open problem, overview existing partial results, and discuss related challenges., Comment: Open problem track. Conference on Learning Theory (COLT 2024)

In the construction sector, ensuring worker safety is of the utmost significance. In this study, a deep learning-based technique is presented for identifying safety gear worn by construction workers, such as helmets, goggles, jackets, gloves, and footwears. The recommended approach uses the YOLO v7 (You Only Look Once) object detection algorithm to precisely locate these safety items. The dataset utilized in this work consists of labeled images split into training, testing and validation sets. Each image has bounding box labels that indicate where the safety equipment is located within the image. The model is trained to identify and categorize the safety equipment based on the labeled dataset through an iterative training approach. We used custom dataset to train this model. Our trained model performed admirably well, with good precision, recall, and F1-score for safety equipment recognition. Also, the model's evaluation produced encouraging results, with a mAP@0.5 score of 87.7\%. The model performs effectively, making it possible to quickly identify safety equipment violations on building sites. A thorough evaluation of the outcomes reveals the model's advantages and points up potential areas for development. By offering an automatic and trustworthy method for safety equipment detection, this research makes a contribution to the fields of computer vision and workplace safety. The proposed deep learning-based approach will increase safety compliance and reduce the risk of accidents in the construction industry

Modern data-driven control applications call for flexible nonlinear models that are amenable to principled controller synthesis and realtime feedback. Many nonlinear dynamical systems of interest are control affine. We propose two novel classes of nonlinear feature representations which capture control affine structure while allowing for arbitrary complexity in the state dependence. Our methods make use of random features (RF) approximations, inheriting the expressiveness of kernel methods at a lower computational cost. We formalize the representational capabilities of our methods by showing their relationship to the Affine Dot Product (ADP) kernel proposed by Casta\~neda et al. (2021) and a novel Affine Dense (AD) kernel that we introduce. We further illustrate the utility by presenting a case study of data-driven optimization-based control using control certificate functions (CCF). Simulation experiments on a double pendulum empirically demonstrate the advantages of our methods., Comment: 25 pages, 3 figures

Dynamic multi-relational graphs are an expressive relational representation for data enclosing entities and relations of different types, and where relationships are allowed to vary in time. Addressing predictive tasks over such data requires the ability to find structure embeddings that capture the diversity of the relationships involved, as well as their dynamic evolution. In this work, we establish a novel class of challenging tasks for dynamic multi-relational graphs involving out-of-domain link prediction, where the relationship being predicted is not available in the input graph. We then introduce a novel Graph Neural Network model, named GOOD, designed specifically to tackle the out-of-domain generalization problem. GOOD introduces a novel design concept for multi-relation embedding aggregation, based on the idea that good representations are such when it is possible to disentangle the mixing proportions of the different relational embeddings that have produced it. We also propose five benchmarks based on two retail domains, where we show that GOOD can effectively generalize predictions out of known relationship types and achieve state-of-the-art results. Most importantly, we provide insights into problems where out-of-domain prediction might be preferred to an in-domain formulation, that is, where the relationship to be predicted has very few positive examples., Comment: 8 pages, 3 figures, 3 Tables, conference [accepted in IEEE WCCI 2024]

In the exascale era in which application behavior has large power & energy footprints, per-application job-level awareness of such impression is crucial in taking steps towards achieving efficiency goals beyond performance, such as energy efficiency, and sustainability. To achieve these goals, we have developed a novel low-latency job power profiling machine learning pipeline that can group job-level power profiles based on their shapes as they complete. This pipeline leverages a comprehensive feature extraction and clustering pipeline powered by a generative adversarial network (GAN) model to handle the feature-rich time series of job-level power measurements. The output is then used to train a classification model that can predict whether an incoming job power profile is similar to a known group of profiles or is completely new. With extensive evaluations, we demonstrate the effectiveness of each component in our pipeline. Also, we provide a preliminary analysis of the resulting clusters that depict the power profile landscape of the Summit supercomputer from more than 60K jobs sampled from the year 2021.

The use of autonomous underwater vehicles (AUVs) to accomplish traditionally challenging and dangerous tasks has proliferated thanks to advances in sensing, navigation, manipulation, and on-board computing technologies. Utilizing AUVs in underwater human-robot interaction (UHRI) has witnessed comparatively smaller levels of growth due to limitations in bi-directional communication and significant technical hurdles to bridge the gap between analogies with terrestrial interaction strategies and those that are possible in the underwater domain. A necessary component to support UHRI is establishing a system for safe robotic-diver approach to establish face-to-face communication that considers non-standard human body pose. In this work, we introduce a stereo vision system for enhancing UHRI that utilizes three-dimensional reconstruction from stereo image pairs and machine learning for localizing human joint estimates. We then establish a convention for a coordinate system that encodes the direction the human is facing with respect to the camera coordinate frame. This allows automatic setpoint computation that preserves human body scale and can be used as input to an image-based visual servo control scheme. We show that our setpoint computations tend to agree both quantitatively and qualitatively with experimental setpoint baselines. The methodology introduced shows promise for enhancing UHRI by improving robotic perception of human orientation underwater., Comment: Submitted to the Proceedings of the 2024 IEEE Conference on Robotics & Automation (ICRA)

We report on the experimental realization of Pb1-xSnxTe pentagonal nanowires (NWs) with [110] orientation using molecular beam epitaxy techniques. Using first-principles calculations, we investigate the structural stability in NWs of SnTe and PbTe in three different structural phases: cubic, pentagonal with [001] orientation and pentagonal with [110] orientation. Within a semiclassical approach, we show that the interplay between ionic and covalent bonds favors the formation of pentagonal NWs. Additionally, we find that this pentagonal structure is more likely to occur in tellurides than in selenides. The disclination and twin boundary cause the electronic states originating from the NW core region to generate a conducting band connecting the valence and conduction bands, creating a symmetry-enforced metallic phase. The metallic core band has opposite slopes in the cases of Sn and Te twin boundary, while the bands from the shell are insulating. We finally study the electronic and topological properties of pentagonal NWs unveiling their potential as a new platform for higher-order topology and fractional charge. These pentagonal NWs represent a unique case of intrinsic core-shell one-dimensional nanostructures with distinct structural, electronic and topological properties between the core and the shell region., Comment: 13 pages, 10 figures

We consider regret minimization in a general collaborative multi-agent multi-armed bandit model, in which each agent faces a finite set of arms and may communicate with other agents through a central controller. The optimal arm for each agent in this model is the arm with the largest expected mixed reward, where the mixed reward of each arm is a weighted average of its rewards across all agents, making communication among agents crucial. While near-optimal sample complexities for best arm identification are known under this collaborative model, the question of optimal regret remains open. In this work, we address this problem and propose the first algorithm with order optimal regret bounds under this collaborative bandit model. Furthermore, we show that only a small constant number of expected communication rounds is needed., Comment: Algorithmic Learning Theory (ALT) 2024