Papers by Povilas Daniušis
arXiv (Cornell University), Nov 19, 2023
We propose a simple multivariate normality test based on Kac-Bernstein's characterization, which ... more We propose a simple multivariate normality test based on Kac-Bernstein's characterization, which can be conducted by utilising existing statistical independence tests for sums and differences of data samples. We also perform its empirical investigation, which reveals that for high-dimensional data, the proposed approach may be more efficient than the alternative ones. The accompanying code repository is provided at .
IEEE Access
End-to-end autonomous driving often relies on the concept of learning to imitate from expert demo... more End-to-end autonomous driving often relies on the concept of learning to imitate from expert demonstrations. Since those demonstrations cannot cover all possible variations in data, there always are situations where the trained agents encounter unseen conditions, which results in a shift in the data distribution. One of the most common causes of this shift is changes in weather and lighting conditions. In this study, we suggest using a pre-training based on the visual place recognition (VPR) method, in order to mitigate this effect. We compare the corresponding navigation agent to a baseline agent which relies on the commonly used ImageNet pre-training by evaluating as per the Leaderboard driving benchmark in CARLA environment. According to our experiments, pre-training on the VPR task shows higher resistance to unseen weather conditions. The findings calculated in our study are evaluated over multiple seeds to show statistical consistency. The accompanying open-source code repository can be accessed via https://github.com/Shubhamcl/vpr_pretrained_agent/. INDEX TERMS Imitation learning, autonomous driving, agents, self-driving cars, deep learning, pre-training.
Cornell University - arXiv, Aug 16, 2022
In this paper, we focus on the problem of statistical dependence estimation using characteristic ... more In this paper, we focus on the problem of statistical dependence estimation using characteristic functions. We propose a statistical dependence measure, based on the maximum-norm of the difference between joint and product-marginal characteristic functions. The proposed measure can detect arbitrary statistical dependence between two random vectors of possibly different dimensions, is differentiable, and easily integrable into modern machine learning and deep learning pipelines. We also conduct experiments both with simulated and real data. Our simulations show, that the proposed method can measure statistical dependencies in high-dimensional, non-linear data, and is less affected by the curse of dimensionality, compared to the previous work in this line of research. The experiments with real data demonstrate the potential applicability of our statistical measure for two different empirical inference scenarios, showing statistically significant improvement in the performance characteristics when applied for supervised feature extraction and deep neural network regularization. In addition, we provide a link to the accompanying open-source repository https://bit.ly/3d4ch5I. Preprint. Under review.
ALLSENSORS 2021, The Sixth International Conference on Advances in Sensors, Actuators, Metering and Sensing, Jul 18, 2021
In many important real world applications the initial representation of the data is inconvenient,... more In many important real world applications the initial representation of the data is inconvenient, or even prohibitive for further analysis. For example, in image analysis, text analysis and computational genetics high-dimensional, massive, structural, incomplete, and noisy data sets are common. Therefore, feature extraction, or revelation of informative features from the raw data is one of fundamental machine learning problems. Efficient feature extraction helps to understand data and the process that generates it, reduce costs for future measurements and data analysis. The representation of the structured data as a compact set of informative numeric features allows applying well studied machine learning techniques instead of developing new ones.. The dissertation focuses on supervised and semi-supervised feature extraction methods, which optimize the dependence structure of features. The dependence is measured using the kernel estimator of Hilbert-Schmidt norm of covariance operator (HSIC measure). Two dependence structures are investigated: in the first case we seek features which maximize the dependence on the dependent variable, and in the second one, we additionally minimize the mutual dependence of features. Linear and kernel formulations of HBFE and HSCA are provided. Using Laplacian regularization framework we construct semi-supervised variants of HBFE and HSCA. Suggested algorithms were investigated experimentally using conventional and multilabel classification data sets. The extracted features were classified by k nearest neighbor classifier, and their quality is evaluated by classification performance measures. Experiments show that in certain cases our algorithms are more efficient comparing to PCA or LDA
Intelligent Data Engineering and Automated Learning - IDEAL 2009, 2009
We propose a novel, supervised feature extraction procedure, based on an unbiased estimator of th... more We propose a novel, supervised feature extraction procedure, based on an unbiased estimator of the Hilbert-Schmidt independence criterion (HSIC). The proposed procedure can be directly applied to single-label or multi-label data, also the kernelized version can be applied to any data type, on which a positive definite kernel function has been defined. Computer experiments with various classification data sets reveal that our approach can be applied more efficiently than the alternative ones.
We consider two variables that are related to each other by an invertible function. While it has ... more We consider two variables that are related to each other by an invertible function. While it has previously been shown that the dependence structure of the noise can provide hints to determine which of the two variables is the cause, we presently show that even in the deterministic (noise-free) case, there are asymmetries that can be exploited for causal inference. Our method is based on the idea that if the function and the probability density of the cause are chosen independently, then the distribution of the effect will, in a certain sense, depend on the function. We provide a theoretical analysis of this method, showing that it also works in the low noise regime, and link it to information geometry. We report strong empirical results on various real-world data sets from different domains.
IEEE Instrumentation & Measurement Magazine, 2020
Blind persons need electronic traveling aid (ETA) solutions for better orientation and navigation... more Blind persons need electronic traveling aid (ETA) solutions for better orientation and navigation in unfamiliar indoor environments, with embedded features for detection and recognition of both obstacles and desired destinations such as rooms, staircases, and elevators. Because the use of GPS for locational references is impractical indoors, the development of such navigation systems is challenging and requires a systematic review and evaluation of different technological approaches. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) method, we evaluated and compared current research papers that deal with the prototyping of assistive devices (visual sensory perception substitution with audio and haptic signals) for blind and visually impaired persons. We conducted an instructional assessment of selected indoor navigation prototypes using three main criteria: navigation technologies, sensors, and computer vision approaches. For the latter category, we conducted a separate systematic review, as papers in this research area primarily specialize in software computer vision solutions rather than hardware. The paper provides useful insights for researchers regarding technological instrumentation for the development of ETA solutions for blind and visually impaired (VI) persons in the field of indoor orientation and navigation.
This paper presents a high-level architecture of a computer vision-based system for partial compe... more This paper presents a high-level architecture of a computer vision-based system for partial compensation of lost or impaired human vision. It combines standard smartphone device, external deep learning-based image processing infrastructure and audio/tactile user interface. The proposed architecture is based on input from user-centered design process, involving end-users into system development. The paper discusses user needs and expectations for electronic travelling aids for the blind and highlights limitations of the existing solutions. The suggested architecture may be used as a basis for developing computer vision-based tools for visually impaired individuals. Keywords–computer vision; deep learning; mobile application; aid for blind and visually impaired; audio feedback; tactile feedback; impaired human vision compensation; user-centered design.
Elektronika Ir Elektrotechnika, 2009
In this article we propose a new linear model for regression/classification of matrix input data.... more In this article we propose a new linear model for regression/classification of matrix input data. The algorithm for parameter estimation is constructed, some properties of the model are analyzed. The proposed model was applied for various binary classification problems, experimentally demonstrated, that in the case of small training sample, this model can be more efficient than standard techniques. In each experiment the training sample was selected randomly, the results (correct classification probabilities on the testing set) were averaged, statistical hypothesis about efficiency of the models were tested. By signed rank test most of the results are statistically significant. Bibl. 9 (in English; summaries in English, Russian and Lithuanian).
In this article, we focus on the utilisation of reactive trajectory imitation controllers for goa... more In this article, we focus on the utilisation of reactive trajectory imitation controllers for goal-directed mobile robot navigation. We propose a topological navigation graph (TNG) - an imitation-learning-based framework for navigating through environments with intersecting trajectories. The TNG framework represents the environment as a directed graph composed of deep neural networks. Each vertex of the graph corresponds to a trajectory and is represented by a trajectory identification classifier and a trajectory imitation controller. For trajectory following, we propose the novel use of neural object detection architectures. The edges of TNG correspond to intersections between trajectories and are all represented by a classifier. We provide empirical evaluation of the proposed navigation framework and its components in simulated and real-world environments, demonstrating that TNG allows us to utilise non-goal-directed, imitation-learning methods for goal-directed autonomous navigat...
Assistive Technology, 2020
This survey/interview is a part of a research project titled "Complex research of augmented reali... more This survey/interview is a part of a research project titled "Complex research of augmented reality for the blind and weak-sighted people" (project No. 01.2.2-LMT-K-718-01-0060) funded by European Regional Development. The overall goal of the project is development of a functional computer vision-based travelling aid for the blind and weak-sighted people. This survey aims to identify and describe requirements and expectations visually impaired users have for such technological solutions. The survey is strictly anonymous. The project team is grateful for your time and effort dedicated for filling in this questionnaire/participating in this interview.
Informatica, 2008
In this paper we propose and analyze a multilayer perceptron-like model with matrix inputs. We ap... more In this paper we propose and analyze a multilayer perceptron-like model with matrix inputs. We applied the proposed model to the financial time series prediction problem, compared it with the standard multilayer perceptron model, and got fairly good results.
Lietuvos matematikos rinkinys, 2008
In this paper we propose a kernel-based regression model for matrix patterns (KRMP). The training... more In this paper we propose a kernel-based regression model for matrix patterns (KRMP). The training algorithm is derived. The proposed model was empirically compared with traditional models.
Lietuvos matematikos rinkinys, 2016
We propose a feature extraction algorithm, based on the Hilbert–Schmidt independence criterion (H... more We propose a feature extraction algorithm, based on the Hilbert–Schmidt independence criterion (HSIC) and the maximum dependence – minimum redundancy approach. Experiments with classification data sets demonstrate that suggested Hilbert–Schmidt component analysis (HSCA) algorithm in certain cases may be more efficient than other considered approaches.
Artificial Intelligence, 2012
While conventional approaches to causal inference are mainly based on conditional (in)dependences... more While conventional approaches to causal inference are mainly based on conditional (in)dependences, recent methods also account for the shape of (conditional) distributions. The idea is that the causal hypothesis "X causes Y " imposes that the marginal distribution P X and the conditional distribution P Y |X represent independent mechanisms of nature. Recently it has been postulated that the shortest description of the joint distribution P X,Y should therefore be given by separate descriptions of P X and P Y |X . Since description length in the sense of Kolmogorov complexity is uncomputable, practical implementations rely on other notions of independence. Here we define independence via orthogonality in information space. This way, we can explicitly describe the kind of dependence that occur between P Y and P X|Y making the causal hypothesis "Y causes X" implausible. Remarkably, this asymmetry between cause and effect becomes particularly simple if X and Y are deterministically related. We present an inference method that works in this case. We also discuss some theoretical results for the non-deterministic case although it is not clear how to employ them for a more general inference method.
Uploads
Papers by Povilas Daniušis