arXiv subjects

Bartosz Zieliński

Publications and source records attributed to Bartosz Zieliński.

50 records · Page 3Linked to original sources

arXiv2022

ProPaLL: Probabilistic Partial Label Learning

Partial label learning is a type of weakly supervised learning, where each training instance corresponds to a set of candidate labels, among which only one is true. In this paper, we introduce ProPaLL, a novel probabilistic approach to this problem, which has at least three advantages compared to the existing approaches: it simplifies the training process, improves performance, and can be applied to any deep architecture. Experiments conducted on artificial and real-world datasets indicate that ProPaLL outperforms the existing approaches.

cs.LG↗

arXiv2022

Visual Probing: Cognitive Framework for Explaining Self-Supervised Image Representations

Recently introduced self-supervised methods for image representation learning provide on par or superior results to their fully supervised competitors, yet the corresponding efforts to explain the self-supervised approaches lag behind. Motivated by this observation, we introduce a novel visual probing framework for explaining the self-supervised models by leveraging probing tasks employed previously in natural language processing. The probing tasks require knowledge about semantic relationships between image parts. Hence, we propose a systematic approach to obtain analogs of natural language in vision, such as visual words, context, and taxonomy. Our proposal is grounded in Marr's computational theory of vision and concerns features like textures, shapes, and lines. We show the effectiveness and applicability of those analogs in the context of explaining self-supervised representations. Our key findings emphasize that relations between language and vision can serve as an effective yet intuitive tool for discovering how machine learning models work, independently of data modality. Our work opens a plethora of research pathways towards more explainable and transparent AI.

cs.CV↗

arXiv2021

Classifying bacteria clones using attention-based deep multiple instance learning interpreted by persistence homology

In this work, we analyze if it is possible to distinguish between different clones of the same bacteria species (Klebsiella pneumoniae) based only on microscopic images. It is a challenging task, previously considered impossible due to the high clones similarity. For this purpose, we apply a multi-step algorithm with attention-based multiple instance learning. Except for obtaining accuracy at the level of 0.9, we introduce extensive interpretability based on CellProfiler and persistence homology, increasing the understandability and trust in the model.

cs.CV↗

arXiv2021

Kernel Self-Attention in Deep Multiple Instance Learning

Not all supervised learning problems are described by a pair of a fixed-size input tensor and a label. In some cases, especially in medical image analysis, a label corresponds to a bag of instances (e.g. image patches), and to classify such bag, aggregation of information from all of the instances is needed. There have been several attempts to create a model working with a bag of instances, however, they are assuming that there are no dependencies within the bag and the label is connected to at least one instance. In this work, we introduce Self-Attention Attention-based MIL Pooling (SA-AbMILP) aggregation operation to account for the dependencies between instances. We conduct several experiments on MNIST, histological, microbiological, and retinal databases to show that SA-AbMILP performs better than other models. Additionally, we investigate kernel variations of Self-Attention and their influence on the results.

cs.LG↗

arXiv2020

ProtoPShare: Prototype Sharing for Interpretable Image Classification and Similarity Discovery

In this paper, we introduce ProtoPShare, a self-explained method that incorporates the paradigm of prototypical parts to explain its predictions. The main novelty of the ProtoPShare is its ability to efficiently share prototypical parts between the classes thanks to our data-dependent merge-pruning. Moreover, the prototypes are more consistent and the model is more robust to image perturbations than the state of the art method ProtoPNet. We verify our findings on two datasets, the CUB-200-2011 and the Stanford Cars.

cs.CV↗

arXiv2020

Deep learning approach to description and classification of fungi microscopic images

Diagnosis of fungal infections can rely on microscopic examination, however, in many cases, it does not allow unambiguous identification of the species due to their visual similarity. Therefore, it is usually necessary to use additional biochemical tests. That involves additional costs and extends the identification process up to 10 days. Such a delay in the implementation of targeted treatment is grave in consequences as the mortality rate for immunosuppressed patients is high. In this paper, we apply machine learning approach based on deep learning and bag-of-words to classify microscopic images of various fungi species. Our approach makes the last stage of biochemical identification redundant, shortening the identification process by 2-3 days and reducing the cost of the diagnostic examination.

cs.CV↗

arXiv2019

Persistence Bag-of-Words for Topological Data Analysis

Persistent homology (PH) is a rigorous mathematical theory that provides a robust descriptor of data in the form of persistence diagrams (PDs). PDs exhibit, however, complex structure and are difficult to integrate in today's machine learning workflows. This paper introduces persistence bag-of-words: a novel and stable vectorized representation of PDs that enables the seamless integration with machine learning. Comprehensive experiments show that the new representation achieves state-of-the-art performance and beyond in much less time than alternative approaches.

stat.ML↗

arXiv2019

Processing of missing data by neural networks

We propose a general, theoretically justified mechanism for processing missing data by neural networks. Our idea is to replace typical neuron's response in the first hidden layer by its expected value. This approach can be applied for various types of networks at minimal cost in their modification. Moreover, in contrast to recent approaches, it does not require complete data for training. Experimental results performed on different types of architectures show that our method gives better results than typical imputation strategies and other methods dedicated for incomplete data.

cs.LG↗

arXiv2018

Cascade context encoder for improved inpainting

In this paper, we analyze if cascade usage of the context encoder with increasing input can improve the results of the inpainting. For this purpose, we train context encoder for 64x64 pixels images in a standard way and use its resized output to fill in the missing input region of the 128x128 context encoder, both in training and evaluation phase. As the result, the inpainting is visibly more plausible. In order to thoroughly verify the results, we introduce normalized squared-distortion, a measure for quantitative inpainting evaluation, and we provide its mathematical explanation. This is the first attempt to formalize the inpainting measure, which is based on the properties of latent feature representation, instead of L2 reconstruction loss.

cs.CV↗

arXiv2017

The K-theory of twisted multipullback quantum odd spheres and complex projective spaces

We find multipullback quantum odd-dimensional spheres equipped with natural $U(1)$-actions that yield the multipullback quantum complex projective spaces constructed from Toeplitz cubes as noncommutative quotients. We prove that the noncommutative line bundles associated to multipullback quantum odd spheres are pairwise stably non-isomorphic, and that the $K$-groups of multipullback quantum complex projective spaces and odd spheres coincide with their classical counterparts. We show that these $K$-groups remain the same for more general twisted versions of our quantum odd spheres and complex projective spaces.

math.KT↗

arXiv2016

Topological descriptors for 3D surface analysis

We investigate topological descriptors for 3D surface analysis, i.e. the classification of surfaces according to their geometric fine structure. On a dataset of high-resolution 3D surface reconstructions we compute persistence diagrams for a 2D cubical filtration. In the next step we investigate different topological descriptors and measure their ability to discriminate structurally different 3D surface patches. We evaluate their sensitivity to different parameters and compare the performance of the resulting topological descriptors to alternative (non-topological) descriptors. We present a comprehensive evaluation that shows that topological descriptors are (i) robust, (ii) yield state-of-the-art performance for the task of 3D surface analysis and (iii) improve classification performance when combined with non-topological descriptors.

cs.CV↗

arXiv2014

Piecewise Principal Coactions of Co-Commutative Hopf Algebras

Principal comodule algebras can be thought of as objects representing principal bundles in non-commutative geometry. A crucial component of a principal comodule algebra is a strong connection map. For some applications it suffices to prove that such a map exists, but for others, such as computing the associated bundle projectors or Chern-Galois characters, an explicit formula for a strong connection is necessary. It has been known for some time how to construct a strong connection map on a multi-pullback comodule algebra from strong connections on multi-pullback components, but the known explicit general formula is unwieldy. In this paper we derive a much easier to use strong connection formula, which is not, however, completely general, but is applicable only in the case when a Hopf algebra is co-commutative. Because certain linear splittings of projections in multi-pullback comodule algebras play a crucial role in our construction, we also devote a significant part of the paper to the problem of existence and explicit formulas for such splittings. Finally, we show example application of our work.

math.QA↗

arXiv2014

Nontrivial Deformation of a Trivial Bundle

The ${\rm SU}(2)$-prolongation of the Hopf fibration $S^3\to S^2$ is a trivializable principal ${\rm SU}(2)$-bundle. We present a noncommutative deformation of this bundle to a quantum principal ${\rm SU}_q(2)$-bundle that is not trivializable. On the other hand, we show that the ${\rm SU}_q(2)$-bundle is piecewise trivializable with respect to the closed covering of $S^2$ by two hemispheres intersecting at the equator.

math.QA↗

arXiv2011

Quantum principal bundles over quantum real projective spaces

Two hierarchies of quantum principal bundles over quantum real projective spaces are constructed. One hierarchy contains bundles with U(1) as a structure group, the other has the quantum group $SU_q(2)$ as a fibre. Both hierarchies are obtained by the process of prolongation from bundles with the cyclic group of order 2 as a fibre. The triviality or otherwise of these bundles is determined by using a general criterion for a prolongation of a comodule algebra to be a cleft Hopf-Galois extension.

math.QA↗