Search Results for author: Aapo Hyvarinen

Found 14 papers, 6 papers with code

Nonlinear Independent Component Analysis for Principled Disentanglement in Unsupervised Deep Learning

no code implementations29 Mar 2023 Aapo Hyvarinen, Ilyes Khemakhem, Hiroshi Morioka

A central problem in unsupervised deep learning is how to find useful representations of high-dimensional data, sometimes called "disentanglement".


Optimizing the Noise in Self-Supervised Learning: from Importance Sampling to Noise-Contrastive Estimation

1 code implementation23 Jan 2023 Omar Chehab, Alexandre Gramfort, Aapo Hyvarinen

Nevertheless, we soberly conclude that the optimal noise may be hard to sample from, and the gain in efficiency can be modest compared to choosing the noise distribution equal to the data's.

Self-Supervised Learning

The Optimal Noise in Noise-Contrastive Learning Is Not What You Think

1 code implementation2 Mar 2022 Omar Chehab, Alexandre Gramfort, Aapo Hyvarinen

Learning a parametric model of a data distribution is a well-known statistical problem that has seen renewed interest as it is brought to scale in deep learning.

Contrastive Learning

Autoregressive flow-based causal discovery and inference

2 code implementations18 Jul 2020 Ricardo Pio Monti, Ilyes Khemakhem, Aapo Hyvarinen

We posit that autoregressive flow models are well-suited to performing a range of causal inference tasks - ranging from causal discovery to making interventional and counterfactual predictions.

Causal Discovery Causal Inference +1

Information criteria for non-normalized models

no code implementations15 May 2019 Takeru Matsuda, Masatoshi Uehara, Aapo Hyvarinen

However, model selection methods for general non-normalized models have not been proposed so far.

Model Selection

Causal Discovery with General Non-Linear Relationships Using Non-Linear ICA

no code implementations19 Apr 2019 Ricardo Pio Monti, Kun Zhang, Aapo Hyvarinen

We consider the problem of inferring causal relationships between two or more passively observed variables.

Causal Discovery

Neural Empirical Bayes

no code implementations6 Mar 2019 Saeed Saremi, Aapo Hyvarinen

Kernel density is viewed symbolically as $X\rightharpoonup Y$ where the random variable $X$ is smoothed to $Y= X+N(0,\sigma^2 I_d)$, and empirical Bayes is the machinery to denoise in a least-squares sense, which we express as $X \leftharpoondown Y$.

Denoising Density Estimation

Nonlinear ICA Using Auxiliary Variables and Generalized Contrastive Learning

1 code implementation22 May 2018 Aapo Hyvarinen, Hiroaki Sasaki, Richard E. Turner

Here, we propose a general framework for nonlinear ICA, which, as a special case, can make use of temporal structure.

Contrastive Learning Representation Learning +2

Estimation of Non-Normalized Mixture Models and Clustering Using Deep Representation

no code implementations19 May 2018 Takeru Matsuda, Aapo Hyvarinen

Then, based on the observation that conventional classification learning with neural networks is implicitly assuming an exponential family as a generative model, we introduce a method for clustering unlabeled data by estimating a finite mixture of distributions in an exponential family.

Clustering Image Clustering

Unsupervised Feature Extraction by Time-Contrastive Learning and Nonlinear ICA

2 code implementations NeurIPS 2016 Aapo Hyvarinen, Hiroshi Morioka

Nonlinear independent component analysis (ICA) provides an appealing framework for unsupervised feature learning, but the models proposed so far are not identifiable.

Contrastive Learning Time Series +1

A direct method for estimating a causal ordering in a linear non-Gaussian acyclic model

no code implementations9 Aug 2014 Shohei Shimizu, Aapo Hyvarinen, Yoshinobu Kawahara

Structural equation models and Bayesian networks have been widely used to analyze causal relations between continuous variables.

Bridging Information Criteria and Parameter Shrinkage for Model Selection

no code implementations8 Jul 2013 Kun Zhang, Heng Peng, Laiwan Chan, Aapo Hyvarinen

Model selection based on classical information criteria, such as BIC, is generally computationally demanding, but its properties are well studied.

Model Selection

ParceLiNGAM: A causal ordering method robust against latent confounders

no code implementations29 Mar 2013 Tatsuya Tashiro, Shohei Shimizu, Aapo Hyvarinen, Takashi Washio

In this paper, we propose a new algorithm for learning causal orders that is robust against one typical violation of the model assumptions: latent confounders.

Cannot find the paper you are looking for? You can Submit a new open access paper.