Understanding Likelihood of Normalizing Flow and Image Complexity through the Lens of Out-of-Distribution Detection

• URL: http://arxiv.org/abs/2402.10477v1
• Date: Fri, 16 Feb 2024 06:56:59 GMT
• Title: Understanding Likelihood of Normalizing Flow and Image Complexity through the Lens of Out-of-Distribution Detection
• Authors: Genki Osada, Tsubasa Takahashi, Takashi Nishide
• Abstract summary: We propose a hypothesis that less complex images concentrate in high-density regions in the latent space, resulting in a higher likelihood assignment in the Normalizing Flow (NF) We experimentally demonstrate its validity for five NF architectures, concluding that their likelihood is untrustworthy. We show that this problem can be alleviated by treating image complexity as an independent variable.
• Score: 5.279257531335347
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Out-of-distribution (OOD) detection is crucial to safety-critical machine learning applications and has been extensively studied. While recent studies have predominantly focused on classifier-based methods, research on deep generative model (DGM)-based methods have lagged relatively. This disparity may be attributed to a perplexing phenomenon: DGMs often assign higher likelihoods to unknown OOD inputs than to their known training data. This paper focuses on explaining the underlying mechanism of this phenomenon. We propose a hypothesis that less complex images concentrate in high-density regions in the latent space, resulting in a higher likelihood assignment in the Normalizing Flow (NF). We experimentally demonstrate its validity for five NF architectures, concluding that their likelihood is untrustworthy. Additionally, we show that this problem can be alleviated by treating image complexity as an independent variable. Finally, we provide evidence of the potential applicability of our hypothesis in another DGM, PixelCNN++.

Related papers

• Perplexity Trap: PLM-Based Retrievers Overrate Low Perplexity Documents [64.43980129731587]
  We propose a causal-inspired inference-time debiasing method called Causal Diagnosis and Correction (CDC) CDC first diagnoses the bias effect of the perplexity and then separates the bias effect from the overall relevance score. Experimental results across three domains demonstrate the superior debiasing effectiveness.
  arXiv  Detail & Related papers  (2025-03-11T17:59:00Z)
• Causal Diffusion Autoencoders: Toward Counterfactual Generation via Diffusion Probabilistic Models [17.124075103464392]
  Diffusion models (DPMs) have become the state-of-the-art in high-quality image generation. DPMs have an arbitrary noisy latent space with no interpretable or controllable semantics. We propose CausalDiffAE, a diffusion-based causal representation learning framework to enable counterfactual generation.
  arXiv  Detail & Related papers  (2024-04-27T00:09:26Z)
• A Versatile Causal Discovery Framework to Allow Causally-Related Hidden Variables [28.51579090194802]
  We introduce a novel framework for causal discovery that accommodates the presence of causally-related hidden variables almost everywhere in the causal network. We develop a Rank-based Latent Causal Discovery algorithm, RLCD, that can efficiently locate hidden variables, determine their cardinalities, and discover the entire causal structure over both measured and hidden ones. Experimental results on both synthetic and real-world personality data sets demonstrate the efficacy of the proposed approach in finite-sample cases.
  arXiv  Detail & Related papers  (2023-12-18T07:57:39Z)
• Assessing the capacity of a denoising diffusion probabilistic model to reproduce spatial context [7.289988602420457]
  Diffusion probabilistic models (DDPMs) demonstrate superior image synthesis performance as compared to generative adversarial networks (GANs) These claims have been evaluated using either ensemble-based methods designed for natural images, or conventional measures of image quality such as structural similarity. The studies reveal new and important insights regarding the capacity of DDPMs to learn spatial context.
  arXiv  Detail & Related papers  (2023-09-19T17:58:35Z)
• Nonlinearity, Feedback and Uniform Consistency in Causal Structural Learning [0.8158530638728501]
  Causal Discovery aims to find automated search methods for learning causal structures from observational data. This thesis focuses on two questions in causal discovery: (i) providing an alternative definition of k-Triangle Faithfulness that (i) is weaker than strong faithfulness when applied to the Gaussian family of distributions, and (ii) under the assumption that the modified version of Strong Faithfulness holds.
  arXiv  Detail & Related papers  (2023-08-15T01:23:42Z)
• Causality-Based Multivariate Time Series Anomaly Detection [63.799474860969156]
  We formulate the anomaly detection problem from a causal perspective and view anomalies as instances that do not follow the regular causal mechanism to generate the multivariate data. We then propose a causality-based anomaly detection approach, which first learns the causal structure from data and then infers whether an instance is an anomaly relative to the local causal mechanism. We evaluate our approach with both simulated and public datasets as well as a case study on real-world AIOps applications.
  arXiv  Detail & Related papers  (2022-06-30T06:00:13Z)
• Weakly Supervised Representation Learning with Sparse Perturbations [82.39171485023276]
  We show that if one has weak supervision from observations generated by sparse perturbations of the latent variables, identification is achievable under unknown continuous latent distributions. We propose a natural estimation procedure based on this theory and illustrate it on low-dimensional synthetic and image-based experiments.
  arXiv  Detail & Related papers  (2022-06-02T15:30:07Z)
• Principled Knowledge Extrapolation with GANs [92.62635018136476]
  We study counterfactual synthesis from a new perspective of knowledge extrapolation. We show that an adversarial game with a closed-form discriminator can be used to address the knowledge extrapolation problem. Our method enjoys both elegant theoretical guarantees and superior performance in many scenarios.
  arXiv  Detail & Related papers  (2022-05-21T08:39:42Z)
• Provable RL with Exogenous Distractors via Multistep Inverse Dynamics [85.52408288789164]
  Real-world applications of reinforcement learning (RL) require the agent to deal with high-dimensional observations such as those generated from a megapixel camera. Prior work has addressed such problems with representation learning, through which the agent can provably extract endogenous, latent state information from raw observations. However, such approaches can fail in the presence of temporally correlated noise in the observations.
  arXiv  Detail & Related papers  (2021-10-17T15:21:27Z)
• Discovering Latent Causal Variables via Mechanism Sparsity: A New Principle for Nonlinear ICA [81.4991350761909]
  Independent component analysis (ICA) refers to an ensemble of methods which formalize this goal and provide estimation procedure for practical application. We show that the latent variables can be recovered up to a permutation if one regularizes the latent mechanisms to be sparse.
  arXiv  Detail & Related papers  (2021-07-21T14:22:14Z)
• The Hidden Uncertainty in a Neural Networks Activations [105.4223982696279]
  The distribution of a neural network's latent representations has been successfully used to detect out-of-distribution (OOD) data. This work investigates whether this distribution correlates with a model's epistemic uncertainty, thus indicating its ability to generalise to novel inputs.
  arXiv  Detail & Related papers  (2020-12-05T17:30:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.