In order for language models to aid physics research, they must first encode representations of mathematical and natural language discourse which lead to coherent explanations, with correct ordering and relevance of statements. We present a collection of datasets developed to evaluate the performance of language models in this regard, which measure capabilities with respect to sentence ordering, position, section prediction, and discourse coherence. Analysis of the data reveals equations and sub-disciplines which are most common in physics discourse, as well as the sentence-level frequency of equations and expressions. We present baselines that demonstrate how contemporary language models are challenged by coherence related tasks in physics, even when trained on mathematical natural language objectives.
With quadrotors becoming immensely popular in applications such as relief operations, infrastructure maintenance etc., a key control design challenge arises when the quadrotor has to manoeuvre through constrained spaces during various operational scenarios: for example, inspecting a pipeline within predefined velocity and space, dropping relief material at a precise location under tight spaces etc., under the face of parametric uncertainties and external disturbances. To tackle such scenarios, a controller needs to ensure a predefined tracking accuracy so as not to violate the constraints while simultaneously tackling uncertainties and disturbances. However, state-of-the-art controllers dealing with constrained system motion are either not applicable for an underactuated system like quadrotor, or cannot tackle system uncertainties under full state constraints. This work attempts to fill such a gap in literature by designing Barrier Lyapunov Function (BLF) based robust controllers to satisfy multiple state-constraints while simultaneously negotiating parametric uncertainties and external disturbances. The superiority of the BLF control method over a typical unconstrained controller is demonstrated, followed by a robust control design to satisfy position and orientation constraints on quadrotor dynamics. Finally, full state-constraints on a quadrotor(i.e., constraints on the position, orientation, linear velocity and angular velocity) are satisfied with robust control. For each control design, the closed-loop system stability is studied analytically and the efficacy of the design is validated extensively either via realistic simulation scenarios or via experiments performed on a real quadrotor.
Most Gaze estimation research only works on a setup condition that a camera perfectly captures eyes gaze. They have not literarily specified how to set up a camera correctly for a given position of a person. In this paper, we carry out a study on gaze estimation with a logical camera setup position. We further bring our research in a practical application by using inexpensive edge devices with a realistic scenario. That is, we first set up a shopping environment where we want to grasp customers gazing behaviors. This setup needs an optimal camera position in order to maintain estimation accuracy from existing gaze estimation research. We then apply the state-of-the-art of few-shot learning gaze estimation to reduce training sampling in the inference phase. In the experiment, we perform our implemented research on NVIDIA Jetson TX2 and achieve a reasonable speed, 12 FPS which is faster compared with our reference work, without much degradation of gaze estimation accuracy. The source code is released at https://github.com/linh-gist/GazeEstimationTX2.
Classification of long sequential data is an important Machine Learning task and appears in many application scenarios. Recurrent Neural Networks, Transformers, and Convolutional Neural Networks are three major techniques for learning from sequential data. Among these methods, Temporal Convolutional Networks (TCNs) which are scalable to very long sequences have achieved remarkable progress in time series regression. However, the performance of TCNs for sequence classification is not satisfactory because they use a skewed connection protocol and output classes at the last position. Such asymmetry restricts their performance for classification which depends on the whole sequence. In this work, we propose a symmetric multi-scale architecture called Circular Dilated Convolutional Neural Network (CDIL-CNN), where every position has an equal chance to receive information from other positions at the previous layers. Our model gives classification logits in all positions, and we can apply a simple ensemble learning to achieve a better decision. We have tested CDIL-CNN on various long sequential datasets. The experimental results show that our method has superior performance over many state-of-the-art approaches.
Position-controlled systems driving repetitive tasks are of significant importance in industrial machinery. The electric actuators used in these systems are responsible for a large part of the global energy consumption, indicating that major savings can be made in this field. In this context, motion profile optimization is a very cost-effective solution as it allows for more energy-efficient machines without additional hardware investments or adaptions. In particular, mono-actuated mechanisms with position-dependent system properties have received considerable attention in literature. However, the current state-of-the-art methods often use unbounded design parameters to describe the motion profile. This both increases the computational complexity and hampers the search for a global optimum. In this paper, Chebyshev polynomials are used to describe the motion profile. Moreover, the exact bounds on the Chebyshev design parameters are derived. This both seriously reduces the computational complexity and limits the design space, allowing the application of a global optimizer such as the genetic algorithm. Experiments validate the added value of the chosen approach. In this study, it is found that the energy consumption can be reduced by 62.9% compared to a standard reference motion profile.
Online learning to rank (OLTR) aims to learn a ranker directly from implicit feedback derived from users' interactions, such as clicks. Clicks however are a biased signal: specifically, top-ranked documents are likely to attract more clicks than documents down the ranking (position bias). In this paper, we propose a novel learning algorithm for OLTR that uses reinforcement learning to optimize rankers: Reinforcement Online Learning to Rank (ROLTR). In ROLTR, the gradients of the ranker are estimated based on the rewards assigned to clicked and unclicked documents. In order to de-bias the users' position bias contained in the reward signals, we introduce unbiased reward shaping functions that exploit inverse propensity scoring for clicked and unclicked documents. The fact that our method can also model unclicked documents provides a further advantage in that less users interactions are required to effectively train a ranker, thus providing gains in efficiency. Empirical evaluation on standard OLTR datasets shows that ROLTR achieves state-of-the-art performance, and provides significantly better user experience than other OLTR approaches. To facilitate the reproducibility of our experiments, we make all experiment code available at https://github.com/ielab/OLTR.
This paper investigates the problem of communication relay establishment for multiple agent-based mobile units using a relay vehicle. The objective is to drive autonomously the relay vehicle to attain a position for communication relay establishment while maintaining the other vehicles inside of its field-of-view. A bearing-based control law is proposed for the relay drone and designed for both single and multiple agents. We also provide a collision avoidance scheme that ensures no collisions between the relay and other agents. Numerical simulations and experimental results are reported as well to show the efficacy of the proposed approach.
The rapid mutation of the influenza virus threatens public health. Reassortment among viruses with different hosts can lead to a fatal pandemic. However, it is difficult to detect the original host of the virus during or after an outbreak as influenza viruses can circulate between different species. Therefore, early and rapid detection of the viral host would help reduce the further spread of the virus. We use various machine learning models with features derived from the position-specific scoring matrix (PSSM) and features learned from word embedding and word encoding to infer the origin host of viruses. The results show that the performance of the PSSM-based model reaches the MCC around 95%, and the F1 around 96%. The MCC obtained using the model with word embedding is around 96%, and the F1 is around 97%.
Issues with tracking, precision pointing, and Doppler shift are the major sources of performance loss in laser inter-satellite communication that can severely decrease the coverage and overall performance of satellite constellations. As a solution to these problems, we propose a cooperation strategy in which a high altitude platform station (HAPS) staying at a quasi-stationary position contributes to the inter-satellite connectivity. In this setup, the HAPS node uses two different scheduling approaches: one that relies on the zenith angle; the other on instantaneous signal-to-noise ratio. To quantify the performance of the proposed scheme, overall outage probabilities for the two scheduling methods are obtained. In addition, guidelines for the design of practical inter-satellite networks are provided.
We have developed a technique combining the accuracy of quantum Monte Carlo in describing the electron correlation with the efficiency of a Machine Learning Potential (MLP). We use kernel regression in combination with SOAP (Smooth Overlap of Atomic Position) features, implemented here in a very efficient way. The key ingredients are: i) a sparsification technique, based on farthest point sampling, ensuring generality and transferability of our MLPs and ii) the so called $\Delta$-learning, allowing a small training data set, a fundamental property for highly accurate but computationally demanding calculations, such as the ones based on quantum Monte Carlo. As the first application we present a benchmark study of the liquid-liquid transition of high-pressure hydrogen and show the quality of our MLP, by emphasizing the importance of high accuracy for this very debated subject, where experiments are difficult in the lab, and theory is still far from being conclusive.
Traditional ranking systems are expected to sort items in the order of their relevance and thereby maximize their utility. In fair ranking, utility is complemented with fairness as an optimization goal. Recent work on fair ranking focuses on developing algorithms to optimize for fairness, given position-based exposure. In contrast, we identify the potential of outliers in a ranking to influence exposure and thereby negatively impact fairness. An outlier in a list of items can alter the examination probabilities, which can lead to different distributions of attention, compared to position-based exposure. We formalize outlierness in a ranking, show that outliers are present in realistic datasets, and present the results of an eye-tracking study, showing that users scanning order and the exposure of items are influenced by the presence of outliers. We then introduce OMIT, a method for fair ranking in the presence of outliers. Given an outlier detection method, OMIT improves fair allocation of exposure by suppressing outliers in the top-k ranking. Using an academic search dataset, we show that outlierness optimization leads to a fairer policy that displays fewer outliers in the top-k, while maintaining a reasonable trade-off between fairness and utility.
Systems of small distributed satellites in low Earth orbit (LEO) transmitting cooperatively to a multiple antenna ground station (GS) are investigated. These satellite swarms have the benefit of much higher spatial separation in the transmit antennas than traditional big satellites with antenna arrays, promising a massive increase in spectral efficiency. However, this would require instantaneous perfect channel state information (CSI) and strong cooperation between satellites. In practice, orbital velocities around 7.5 km/s lead to very short channel coherence times on the order of fractions of the inter-satellite propagation delay, invalidating these assumptions. In this paper, we propose a distributed linear precoding scheme and a GS equalizer relying on local position information. In particular, each satellite only requires information about its own position and that of the GS, while the GS has complete positional information. Due to the deterministic nature of satellite movement this information is easily obtained and no inter-satellite information exchange is required during transmission. Based on the underlying geometrical channel approximation, the optimal inter-satellite distance is obtained analytically. Numerical evaluations show that the proposed scheme is, on average, within 99.8% of the maximum achievable rate for instantaneous CSI and perfect cooperation
We present the Brown Pedestrian Odometry Dataset (BPOD) for benchmarking visual odometry algorithms in head-mounted pedestrian settings. This dataset was captured using synchronized global and rolling shutter stereo cameras in 12 diverse indoor and outdoor locations on Brown University's campus. Compared to existing datasets, BPOD contains more image blur and self-rotation, which are common in pedestrian odometry but rare elsewhere. Ground-truth trajectories are generated from stick-on markers placed along the pedestrian's path, and the pedestrian's position is documented using a third-person video. We evaluate the performance of representative direct, feature-based, and learning-based VO methods on BPOD. Our results show that significant development is needed to successfully capture pedestrian trajectories. The link to the dataset is here: \url{https://doi.org/10.26300/c1n7-7p93
Emotion recognition in conversation (ERC) has attracted much attention in recent years for its necessity in widespread applications. Existing ERC methods mostly model the self and inter-speaker context separately, posing a major issue for lacking enough interaction between them. In this paper, we propose a novel Speaker and Position-Aware Graph neural network model for ERC (S+PAGE), which contains three stages to combine the benefits of both Transformer and relational graph convolution network (R-GCN) for better contextual modeling. Firstly, a two-stream conversational Transformer is presented to extract the coarse self and inter-speaker contextual features for each utterance. Then, a speaker and position-aware conversation graph is constructed, and we propose an enhanced R-GCN model, called PAG, to refine the coarse features guided by a relative positional encoding. Finally, both of the features from the former two stages are input into a conditional random field layer to model the emotion transfer.
The unmanned aerial vehicle (UAV) is one of the technological breakthroughs that supports a variety of services, including communications. UAV will play a critical role in enhancing the physical layer security of wireless networks. This paper defines the problem of eavesdropping on the link between the ground user and the UAV, which serves as an aerial base station (ABS). The reinforcement learning algorithms Q-learning and deep Q-network (DQN) are proposed for optimizing the position of the ABS and the transmission power to enhance the data rate of the ground user. This increases the secrecy capacity without the system knowing the location of the eavesdropper. Simulation results show fast convergence and the highest secrecy capacity of the proposed DQN compared to Q-learning and baseline approaches.
Active visual tracking of space non-cooperative object is significant for future intelligent spacecraft to realise space debris removal, asteroid exploration, autonomous rendezvous and docking. However, existing works often consider this task into different subproblems (e.g. image preprocessing, feature extraction and matching, position and pose estimation, control law design) and optimize each module alone, which are trivial and sub-optimal. To this end, we propose an end-to-end active visual tracking method based on DQN algorithm, named as DRLAVT. It can guide the chasing spacecraft approach to arbitrary space non-cooperative target merely relied on color or RGBD images, which significantly outperforms position-based visual servoing baseline algorithm that adopts state-of-the-art 2D monocular tracker, SiamRPN. Extensive experiments implemented with diverse network architectures, different perturbations and multiple targets demonstrate the advancement and robustness of DRLAVT. In addition, We further prove our method indeed learnt the motion patterns of target with deep reinforcement learning through hundreds of trial-and-errors.
Deep learning (DL) models are very effective on many computer vision problems and increasingly used in critical applications. They are also inherently black box. A number of methods exist to generate image-wise explanations that allow practitioners to understand and verify model predictions for a given image. Beyond that, it would be desirable to validate that a DL model \textit{generally} works in a sensible way, i.e. consistent with domain knowledge and not relying on undesirable data artefacts. For this purpose, the model needs to be explained globally. In this work, we focus on image modalities that are naturally aligned such that each pixel position represents a similar relative position on the imaged object, as is common in medical imaging. We propose the pixel-wise aggregation of image-wise explanations as a simple method to obtain label-wise and overall global explanations. These can then be used for model validation, knowledge discovery, and as an efficient way to communicate qualitative conclusions drawn from inspecting image-wise explanations. We further propose Progressive Erasing Plus Progressive Restoration (PEPPR) as a method to quantitatively validate that these global explanations are faithful to how the model makes its predictions. We then apply these methods to ultra-widefield retinal images, a naturally aligned modality. We find that the global explanations are consistent with domain knowledge and faithfully reflect the model's workings.
Indoor positioning systems (IPSs) can make an important contribution to the analysis and optimization of internal transport processes. The overall aim of this work is to examine how position data can be used to analyze the operation of industrial trucks in warehouses. This is achieved by developing a concept for the analysis of industrial truck operations based merely on position data. The concept consists of a signal processing scheme to derive kinematic data and three analysis methods - Monitoring, Area analysis, and Motion analysis. Schemes for the signal processing and detection of motion events were developed and implemented as part of the TrOpLocerApp (Truck Operation Localization Analyzer-Application) for recording, displaying, and processing position data, according to the proposed system concept. The TrOpLocer-App source code is published on GitLab. Different filter algorithms were examined, as part of the signal processing scheme, from which the low pass Butterworth filter has shown the best results in static experiments. Validation of the motion detection scheme shows good detection quality for distinct events in a realistic movement experiment.
We present a numerical method to model dynamical systems from data. We use the recently introduced method Scalable Probabilistic Approximation (SPA) to project points from a Euclidean space to convex polytopes and represent these projected states of a system in new, lower-dimensional coordinates denoting their position in the polytope. We then introduce a specific nonlinear transformation to construct a model of the dynamics in the polytope and to transform back into the original state space. To overcome the potential loss of information from the projection to a lower-dimensional polytope, we use memory in the sense of the delay-embedding theorem of Takens. By construction, our method produces stable models. We illustrate the capacity of the method to reproduce even chaotic dynamics and attractors with multiple connected components on various examples.
Deep neural networks (DNNs) have been broadly adopted in health risk prediction to provide healthcare diagnoses and treatments. To evaluate their robustness, existing research conducts adversarial attacks in the white/gray-box setting where model parameters are accessible. However, a more realistic black-box adversarial attack is ignored even though most real-world models are trained with private data and released as black-box services on the cloud. To fill this gap, we propose the first black-box adversarial attack method against health risk prediction models named MedAttacker to investigate their vulnerability. MedAttacker addresses the challenges brought by EHR data via two steps: hierarchical position selection which selects the attacked positions in a reinforcement learning (RL) framework and substitute selection which identifies substitute with a score-based principle. Particularly, by considering the temporal context inside EHRs, it initializes its RL position selection policy by using the contribution score of each visit and the saliency score of each code, which can be well integrated with the deterministic substitute selection process decided by the score changes. In experiments, MedAttacker consistently achieves the highest average success rate and even outperforms a recent white-box EHR adversarial attack technique in certain cases when attacking three advanced health risk prediction models in the black-box setting across multiple real-world datasets. In addition, based on the experiment results we include a discussion on defending EHR adversarial attacks.