This paper introduces the Pareto Data Framework, an approach for identifying and selecting the Minimum Viable Data (MVD) required for enabling machine learning applications on constrained platforms such as embedded systems, mobile devices, and Internet of Things (IoT) devices. We demonstrate that strategic data reduction can maintain high performance while significantly reducing bandwidth, energy, computation, and storage costs. The framework identifies Minimum Viable Data (MVD) to optimize efficiency across resource-constrained environments without sacrificing performance. It addresses common inefficient practices in an IoT application such as overprovisioning of sensors and overprecision, and oversampling of signals, proposing scalable solutions for optimal sensor selection, signal extraction and transmission, and data representation. An experimental methodology demonstrates effective acoustic data characterization after downsampling, quantization, and truncation to simulate reduced-fidelity sensors and network and storage constraints; results shows that performance can be maintained up to 95\% with sample rates reduced by 75\% and bit depths and clip length reduced by 50\% which translates into substantial cost and resource reduction. These findings have implications on the design and development of constrained systems. The paper also discusses broader implications of the framework, including the potential to democratize advanced AI technologies across IoT applications and sectors such as agriculture, transportation, and manufacturing to improve access and multiply the benefits of data-driven insights.

Model Predictive Control (MPC) is a common tool for the control of nonlinear, real-world systems, such as legged robots. However, solving MPC quickly enough to enable its use in real-time is often challenging. One common solution is given by real-time iterations, which does not solve the MPC problem to convergence, but rather close enough to give an approximate solution. In this paper, we extend this idea to a bilevel control framework where a "high-level" optimization program modifies a controller parameter of a "low-level" MPC problem which generates the control inputs and desired state trajectory. We propose an algorithm to iterate on this bilevel program in real-time and provide conditions for its convergence and improvements in stability. We then demonstrate the efficacy of this algorithm by applying it to a quadrupedal robot where the high-level problem optimizes a contact schedule in real-time. We show through simulation that the algorithm can yield improvements in disturbance rejection and optimality, while creating qualitatively new gaits.

Abstract: In the field of global energy and environment, crude oil is an important strategic resource, and its price fluctuation has a far-reaching impact on the global economy, financial market and the process of low-carbon development. In recent years, with the gradual promotion of green energy transformation and low-carbon development in various countries, the dynamics of crude oil market have become more complicated and changeable. The price of crude oil is not only influenced by traditional factors such as supply and demand, geopolitical conflict and production technology, but also faces the challenges of energy policy transformation, carbon emission control and new energy technology development. This diversified driving factor makes the prediction of crude oil price not only very important in economic decision-making and energy planning, but also a key issue in financial markets.In this paper, the spot price data of European Brent crude oil provided by us energy information administration are selected, and a deep learning model with three layers of LSTM units is constructed to predict the crude oil price in the next few days. The results show that the LSTM model performs well in capturing the overall price trend, although there is some deviation during the period of sharp price fluctuation. The research in this paper not only verifies the applicability of LSTM model in energy market forecasting, but also provides data support for policy makers and investors when facing the uncertainty of crude oil price.

Combustion is the primary process in gas turbine engines, where there is a need for efficient air-fuel mixing to enhance performance. High-shear swirl injectors are commonly used to improve fuel atomization and mixing, which are key factors in determining combustion efficiency and emissions. However, under certain conditions, combustors can experience thermoacoustic instability. In this study, a decision tree-based machine learning algorithm is used to classify combustion conditions by analyzing acoustic pressure and high-speed flame imaging from a counter-rotating high-shear swirl injector of a single can combustor fueled by methane. With a constant Reynolds number and varying equivalence ratios, the combustor exhibits both stable and unstable states. Characteristic features are extracted from the data using time series analysis, providing insight into combustion dynamics. The trained supervised machine learning model accurately classifies stable and unstable operations, demonstrating effective prediction of combustion conditions within the studied parameter range.

Bushfire is one of the major natural disasters that cause huge losses to livelihoods and the environment. Understanding and analyzing the severity of bushfires is crucial for effective management and mitigation strategies, helping to prevent the extensive damage and loss caused by these natural disasters. This study presents an in-depth analysis of bushfire severity in Australia over the last twelve years, combining remote sensing data and machine learning techniques to predict future fire trends. By utilizing Landsat imagery and integrating spectral indices like NDVI, NBR, and Burn Index, along with topographical and climatic factors, we developed a robust predictive model using XGBoost. The model achieved high accuracy, 86.13%, demonstrating its effectiveness in predicting fire severity across diverse Australian ecosystems. By analyzing historical trends and integrating factors such as population density and vegetation cover, we identify areas at high risk of future severe bushfires. Additionally, this research identifies key regions at risk, providing data-driven recommendations for targeted firefighting efforts. The findings contribute valuable insights into fire management strategies, enhancing resilience to future fire events in Australia. Also, we propose future work on developing a UAV-based swarm coordination model to enhance fire prediction in real-time and firefighting capabilities in the most vulnerable regions.

Dynamic topic modeling is useful at discovering the development and change in latent topics over time. However, present methodology relies on algorithms that separate document and word representations. This prevents the creation of a meaningful embedding space where changes in word usage and documents can be directly analyzed in a temporal context. This paper proposes an expansion of the compass-aligned temporal Word2Vec methodology into dynamic topic modeling. Such a method allows for the direct comparison of word and document embeddings across time in dynamic topics. This enables the creation of visualizations that incorporate temporal word embeddings within the context of documents into topic visualizations. In experiments against the current state-of-the-art, our proposed method demonstrates overall competitive performance in topic relevancy and diversity across temporal datasets of varying size. Simultaneously, it provides insightful visualizations focused on temporal word embeddings while maintaining the insights provided by global topic evolution, advancing our understanding of how topics evolve over time.

Traditional control methods effectively manage robot operations using models like motion equations but face challenges with issues of contact and friction, leading to unstable and imprecise controllers that often require manual tweaking. Reinforcement learning, however, has developed as a capable solution for developing robust robot controllers that excel in handling contact-related challenges. In this work, we introduce a deep reinforcement learning approach to tackle the soft-capture phase for free-floating moving targets, mainly space debris, amidst noisy data. Our findings underscore the crucial role of tactile sensors, even during the soft-capturing phase. By employing deep reinforcement learning, we eliminate the need for manual feature design, simplifying the problem and allowing the robot to learn soft-capture strategies through trial and error. To facilitate effective learning of the approach phase, we have crafted a specialized reward function that offers clear and insightful feedback to the agent. Our method is trained entirely within the simulation environment, eliminating the need for direct demonstrations or prior knowledge of the task. The developed control policy shows promising results, highlighting the necessity of using tactile sensor information. The code and simulation results are available at Soft_Capture_Tactile repo.

Both computational and experimental material discovery bring forth the challenge of exploring multidimensional and multimodal parameter spaces, such as phase diagrams of Hamiltonians with multiple interactions, composition spaces of combinatorial libraries, material structure image spaces, and molecular embedding spaces. Often these systems are black-box and time-consuming to evaluate, which resulted in strong interest towards active learning methods such as Bayesian optimization (BO). However, these systems are often noisy which make the black box function severely multi-modal and non-differentiable, where a vanilla BO can get overly focused near a single or faux optimum, deviating from the broader goal of scientific discovery. To address these limitations, here we developed Strategic Autonomous Non-Smooth Exploration (SANE) to facilitate an intelligent Bayesian optimized navigation with a proposed cost-driven probabilistic acquisition function to find multiple global and local optimal regions, avoiding the tendency to becoming trapped in a single optimum. To distinguish between a true and false optimal region due to noisy experimental measurements, a human (domain) knowledge driven dynamic surrogate gate is integrated with SANE. We implemented the gate-SANE into a pre-acquired Piezoresponse spectroscopy data of a ferroelectric combinatorial library with high noise levels in specific regions, and a piezoresponse force microscopy (PFM) hyperspectral data. SANE demonstrated better performance than classical BO to facilitate the exploration of multiple optimal regions and thereby prioritized learning with higher coverage of scientific values in autonomous experiments. Our work showcases the potential application of this method to real-world experiment, where such combined strategic and human intervening approaches can be critical to unlocking new discoveries in autonomous research.

Models of digital computation, which lie at the foundation of computer science, are typically discrete, while most of our fundamental models of the physical world are essentially continuous. Nonetheless, the Church-Turing thesis [Tur39] and its physical counterparts [Gan80, CS07] state that this difference is illusory: the discrete computations we can perform reliably in the physical world should be the same as those which can be performed by a Turing machine, possibly by one having access to random bits. The validity of the physical Church-Turing thesis is a subject of debate, and a number of variants of the thesis have been proposed [Cop97]. Furthermore, from the perspective of complexity theory rather than computatibility theory, the possibility for quantum computers to solve with high probability, in polynomial time, decision problems which are not in P, is a basic motivation for research on quantum computation [NC10, ACQ22]. In a different (non-quantum) direction, there have been multiple models proposed for a definition of a computable real function [Grz55, Lac59, Blu98, Sma97, Bra05a], and using this language, it has been found that simple finite-dimensional continuous dynamical systems defined by polynomial equations with integral coefficients can exhibit non-computable dynamical properties [Moo90, BY06]. In general it is known that the existence of natural problems with no computable solution (such as the problem of recognizing presentations of the trivial group [PS]) forces complex behaviour of various continuous mathematical objects related to geometry and dynamics [Wei20, Sei08]. In yet a different direction, there has been a sequence of papers asking whether universal computation can be realized by various ordinary [Bra94] and partial differential equations, including in single-particle potential energy systems [Tao17] and in solutions to fluid dynamics equations [CMPSP21]; this was in part motivated by the hope of showing the existence of blow-up solutions to the Navier-Stokes equations by finding fluid flows which'replicate themselves' at smaller and smaller scales [Tao16]. Such works on realizing universal computation in natural continuous physical models can be seen as a continuation of Moore's earlier work [Moo98, Moo90], which realized universal computation in a simple 2-dimensional piecewise-linear map, as well as in a Lipschitz map on the interval and an analytic map on R. The relation between the computational capacity and the analytic or dynamical properties of a continuous dynamical system, such as its topological entropy or its regularity, are known to be subtle: for example, depending on the formalization, the topological entropy of a Turing-universal system can be zero [CMPS23] or can be forced to be nonnegative [BCMPS24].

Urban intersections, filled with a diverse mix of vehicles from small cars to large semi-trailers, present a persistent challenge for traffic control and management. This reality drives our investigation into how robot vehicles (RVs) can transform such heterogeneous traffic flow, particularly at unsignalized intersections where traditional control methods often falter during power failures and emergencies. Using reinforcement learning (RL) and real-world traffic data, we study heterogeneous mixed traffic across complex intersections under gradual automation by varying RV penetration from 10% to 90%. The results are compelling: average waiting times decrease by up to 86% and 91% compared to signalized and unsignalized intersections, respectively. Additionally, we uncover a "rarity advantage," where less frequent vehicles, such as trucks, benefit the most from RV coordination (by up to 87%). RVs' presence also leads to lower CO2 emissions and fuel consumption compared to managing traffic via traffic lights. Moreover, space headways decrease across all vehicle types as RV rate increases, indicating better road space utilization.