Conditional GANs are frequently used for manipulating the attributes of face images, such as expression, hairstyle, pose, or age.
Prototypical parts-based networks are becoming increasingly popular due to their faithful self-explanations.
Graph Neural Networks (GNNs) play a fundamental role in many deep learning problems, in particular in cheminformatics.
The key components of this framework are the pruning and growing criteria, which are repeatedly applied during the training process to adjust the network's sparse connectivity.
The most common MIL model is when we consider a bag as positive if at least one of its instances has a positive label.
By learning to remain invariant to applied data augmentations, methods such as SimCLR and MoCo are able to reach quality on par with supervised approaches.
Consequently, we can only replace the 2D images (without additional training) to produce a NeRF representation of the new object.
Deep clustering has been dominated by flat models, which split a dataset into a predefined number of groups.
Ranked #1 on Image Clustering on MNIST
Implicit Neural Representations (INRs) are nowadays used to represent multimedia signals across various real-life applications, including image super-resolution, image compression, or 3D rendering.
Recent transformer language models achieve outstanding results in many natural language processing (NLP) tasks.
We introduce ProtoSeg, a novel model for interpretable semantic image segmentation, which constructs its predictions using similar patches from the training set.
Implicit neural representations (INRs) are a rapidly growing research field, which provides alternative ways to represent multimedia signals.
Unfortunately, previous modifications of MAML are limited due to the simplicity of Gaussian posteriors, MAML-like gradient-based weight updates, or by the same structure enforced for universal and adapted weights.
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.
We attempt to address that challenge by proposing a novel approach to the problem: Local Intrinsic Dimension estimation using approximate Likelihood (LIDL).
The proposed classifier is confident only if a single class has a high probability and other probabilities are negligible.
Many crucial problems in deep learning and statistics are caused by a variational gap, i. e., a difference between evidence and evidence lower bound (ELBO).
We introduce a new training paradigm that enforces interval constraints on neural network parameter space to control forgetting.
Few-shot models aim at making predictions using a minimal number of labeled examples from a given task.
We introduce ProtoPool, an interpretable image classification model with a pool of prototypes shared by the classes.
Processing of missing data by modern neural networks, such as CNNs, remains a fundamental, yet unsolved challenge, which naturally arises in many practical applications, like image inpainting or autonomous vehicles and robots.
This makes the GP posterior locally non-Gaussian, therefore we name our method Non-Gaussian Gaussian Processes (NGGPs).
Self-supervised learning holds promise to revolutionize molecule property prediction - a central task to drug discovery and many more industries - by enabling data efficient learning from scarce experimental data.
We propose FlowSVDD -- a flow-based one-class classifier for anomaly/outliers detection that realizes a well-known SVDD principle using deep learning tools.
Recent years have seen a surge in research on deep interpretable neural networks with decision trees as one of the most commonly incorporated tools.
The adoption of Transformer-based models in natural language processing (NLP) has led to great success using a massive number of parameters.
The problem of reducing processing time of large deep learning models is a fundamental challenge in many real-world applications.
In this work, we reformulate the problem of point cloud completion into an object hallucination task.
Predicting future states or actions of a given system remains a fundamental, yet unsolved challenge of intelligence, especially in the scope of complex and non-deterministic scenarios, such as modeling behavior of humans.
In this paper, we introduce ProtoPShare, a self-explained method that incorporates the paradigm of prototypical parts to explain its predictions.
We propose OneFlow - a flow-based one-class classifier for anomaly (outlier) detection that finds a minimal volume bounding region.
We present a mechanism for detecting adversarial examples based on data representations taken from the hidden layers of the target network.
To that end, we devise a generative model that uses a hypernetwork to return the weights of a Continuous Normalizing Flows (CNF) target network.
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.
We develop a fast end-to-end method for training lightweight neural networks using multiple classifier heads.
We argue for the existence of the "break-even" point on this trajectory, beyond which the curvature of the loss surface and noise in the gradient are implicitly regularized by SGD.
Designing a single neural network architecture that performs competitively across a range of molecule property prediction tasks remains largely an open challenge, and its solution may unlock a widespread use of deep learning in the drug discovery industry.
In the paper we construct a fully convolutional GAN model: LocoGAN, which latent space is given by noise-like images of possibly different resolutions.
The main idea of our HyperCloud method is to build a hyper network that returns weights of a particular neural network (target network) trained to map points from a uniform unit ball distribution into a 3D shape.
Graph Convolutional Networks (GCNs) have recently become the primary choice for learning from graph-structured data, superseding hash fingerprints in representing chemical compounds.
We propose a semi-supervised generative model, SeGMA, which learns a joint probability distribution of data and their classes and which is implemented in a typical Wasserstein auto-encoder framework.
We present an efficient technique, which allows to train classification networks which are verifiably robust against norm-bounded adversarial attacks.
Independent Component Analysis (ICA) - one of the basic tools in data analysis - aims to find a coordinate system in which the components of the data are independent.
In the case of geodesics, we search for the curves with the shortest length, while in the case of generative models we typically apply linear interpolation in the latent space.
Motivated by the human way of memorizing images we introduce their functional representation, where an image is represented by a neural network.
In this paper, we propose a simple, fast and easy to implement algorithm LOSSGRAD (locally optimal step-size in gradient descent), which automatically modifies the step-size in gradient descent during neural networks training.
Global pooling, such as max- or sum-pooling, is one of the key ingredients in deep neural networks used for processing images, texts, graphs and other types of structured data.
We construct a general unified framework for learning representation of structured data, i. e. data which cannot be represented as the fixed-length vectors (e. g. sets, graphs, texts or images of varying sizes).
We propose that in ResNets this can be resolved based on our results, by ensuring the same level of dynamical isometry at initialization.
The crucial new ingredient is the introduction of a new (Cramer-Wold) metric in the space of densities, which replaces the Wasserstein metric used in SWAE.
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.
In this paper we propose a mixture model, SparseMix, for clustering of sparse high dimensional binary data, which connects model-based with centroid-based clustering.
In order to graphically represent and interpret the results the notion of Voronoi diagram was generalized to non Euclidean spaces and applied for introduced clustering method.
Incomplete data are often represented as vectors with filled missing attributes joined with flag vectors indicating missing components.
The R Package CEC performs clustering based on the cross-entropy clustering (CEC) method, which was recently developed with the use of information theory.
Representation learning is currently a very hot topic in modern machine learning, mostly due to the great success of the deep learning methods.
The main contribution of this paper is proposing a model based on the information theoretic concepts which on the one hand shows new, entropic perspective on known linear classifiers and on the other leads to a construction of very robust method competetitive with the state of the art non-information theoretic ones (including Support Vector Machines and Extreme Learning Machines).
In the classical Gaussian SVM classification we use the feature space projection transforming points to normal distributions with fixed covariance matrices (identity in the standard RBF and the covariance of the whole dataset in Mahalanobis RBF).
Then we prove that our method is a multithreshold large margin classifier, which shows the analogy to the SVM, while in the same time works with much broader class of hypotheses.