scholarly journals A Survey on Variational Autoencoders from a Green AI Perspective

2021 ◽  
Vol 2 (4) ◽  
Author(s):  
Andrea Asperti ◽  
Davide Evangelista ◽  
Elena Loli Piccolomini
Keyword(s):  
Architectural Design ◽  
Mathematical Formulation ◽  
Representation Learning ◽  
Generative Models ◽  
Model Description ◽  
Energetic Efficiency ◽  
Detailed Model ◽  
Latent Distribution ◽  
Quantitative Results ◽  
Generation Problem

AbstractVariational Autoencoders (VAEs) are powerful generative models that merge elements from statistics and information theory with the flexibility offered by deep neural networks to efficiently solve the generation problem for high-dimensional data. The key insight of VAEs is to learn the latent distribution of data in such a way that new meaningful samples can be generated from it. This approach led to tremendous research and variations in the architectural design of VAEs, nourishing the recent field of research known as unsupervised representation learning. In this article, we provide a comparative evaluation of some of the most successful, recent variations of VAEs. We particularly focus the analysis on the energetic efficiency of the different models, in the spirit of the so-called Green AI, aiming both to reduce the carbon footprint and the financial cost of generative techniques. For each architecture, we provide its mathematical formulation, the ideas underlying its design, a detailed model description, a running implementation and quantitative results.

scholarly journals Variational Autoencoder-Based Multiple Image Captioning Using a Caption Attention Map

2019 ◽  
Vol 9 (13) ◽  
pp. 2699 ◽  
Author(s):  
Boeun Kim ◽  
Saim Shin ◽  
Hyedong Jung
Keyword(s):  
Generative Models ◽  
Research Topic ◽  
Video Data ◽  
Image Feature ◽  
Image Captioning ◽  
Multiple Image ◽  
Visually Impaired People ◽  
Proposed Model ◽  
Variational Autoencoder ◽  
Latent Distribution

Image captioning is a promising research topic that is applicable to services that search for desired content in a large amount of video data and a situation explanation service for visually impaired people. Previous research on image captioning has been focused on generating one caption per image. However, to increase usability in applications, it is necessary to generate several different captions that contain various representations for an image. We propose a method to generate multiple captions using a variational autoencoder, which is one of the generative models. Because an image feature plays an important role when generating captions, a method to extract a Caption Attention Map (CAM) of the image is proposed, and CAMs are projected to a latent distribution. In addition, methods for the evaluation of multiple image captioning tasks are proposed that have not yet been actively researched. The proposed model outperforms in the aspect of diversity compared with the base model when the accuracy is comparable. Moreover, it is verified that the model using CAM generates detailed captions describing various content in the image.

Nanotectonica: Architectural Design Studio and Table Top SEM

2008 ◽  
Vol 16 (5) ◽  
pp. 40-43
Author(s):  
Jonas Coersmeier ◽  
Donovan N. Leonard
Keyword(s):  
Architectural Design ◽  
Generative Models ◽  
Design Studio ◽  
Research Seminar ◽  
Large Span ◽  
Mother Nature ◽  
Macro Scale ◽  
Buckminster Fuller ◽  
Design Students ◽  
Architectural Design Studio

Inspired by architect Frei Otto and design scientist Buckminster Fuller, third year Pratt Institute design students from Jonas Coersmeier’s design studio and research seminar (of Spring 2008) utilized a Table Top SEM to observe micro and nano-scale features produced solely by Mother Nature. After analyzing and documenting the intricacy, beauty and functionality of natural structures, students selected structural entities typically not observed on the macro scale, and utilized the micrograph data to generate analytical drawings followed by generative models for design of a large span structure that would become an aquatic center in the Williamsburg neighborhood of Brooklyn, N.Y.

scholarly journals An Iterative Decomposed Piecewise-Linear Model Description

2009 ◽  
Vol 2009 ◽  
pp. 1-5
Author(s):  
Victor Jimenez-Fernandez ◽  
Luis Hernandez-Martinez ◽  
Arturo Sarmiento-Reyes
Keyword(s):  
Linear Model ◽  
Piecewise Linear ◽  
Mathematical Formulation ◽  
Model Description ◽  
One Dimensional ◽  
Piecewise Linear Model ◽  
Dependent Variables

A model description for the representation of one-dimensional piecewise-linear characteristics is presented. The model can be denoted as a decomposed one, because the independent and dependent variables of the PWL characteristic are treated separately. It is also called iterative, because the particular representation of each segment of the PWL characteristic depends on the value of only one parameter included in the mathematical formulation, it gives the possibility of modeling both, univalued and multivalued PWL characteristics.

scholarly journals TARA: Training and Representation Alteration for AI Fairness and Domain Generalization

2022 ◽  
pp. 1-38
Author(s):  
William Paul ◽  
Armin Hadzic ◽  
Neil Joshi ◽  
Fady Alajaji ◽  
Philippe Burlina
Keyword(s):  
Domain Adaptation ◽  
Representation Learning ◽  
Data Representation ◽  
Generative Models ◽  
Underrepresented Populations ◽  
Latent Space ◽  
Dual Strategy ◽  
Fine Control ◽  
Novel Method ◽  
And Training

Abstract We propose a novel method for enforcing AI fairness with respect to protected or sensitive factors. This method uses a dual strategy performing training and representation alteration (TARA) for the mitigation of prominent causes of AI bias. It includes the use of representation learning alteration via adversarial independence to suppress the bias-inducing dependence of the data representation from protected factors and training set alteration via intelligent augmentation to address bias-causing data imbalance by using generative models that allow the fine control of sensitive factors related to underrepresented populations via domain adaptation and latent space manipulation. When testing our methods on image analytics, experiments demonstrate that TARA significantly or fully debiases baseline models while outperforming competing debiasing methods that have the same amount of information—for example, with (% overall accuracy, % accuracy gap) = (78.8, 0.5) versus the baseline method's score of (71.8, 10.5) for Eye-PACS, and (73.7, 11.8) versus (69.1, 21.7) for CelebA. Furthermore, recognizing certain limitations in current metrics used for assessing debiasing performance, we propose novel conjunctive debiasing metrics. Our experiments also demonstrate the ability of these novel metrics in assessing the Pareto efficiency of the proposed methods.

scholarly journals Guiding Representation Learning in Deep Generative Models with Policy Gradients

2021 ◽  
pp. 115-131
Author(s):  
Luca Lach ◽  
Timo Korthals ◽  
Francesco Ferro ◽  
Helge Ritter ◽  
Malte Schilling
Keyword(s):  
Representation Learning ◽  
Generative Models

scholarly journals Neural generative models and representation learning for information retrieval

2019 ◽  
Vol 53 (2) ◽  
pp. 97-97
Author(s):  
Qingyao Ai
Keyword(s):  
Information Retrieval ◽  
Theoretical Analysis ◽  
Language Processing ◽  
Ad Hoc ◽  
Representation Learning ◽  
Generative Models ◽  
Neural Models ◽  
Retrieval Models ◽  
Types Of Information ◽  
Text Images

Information Retrieval (IR) concerns about the structure, analysis, organization, storage, and retrieval of information. Among different retrieval models proposed in the past decades, generative retrieval models, especially those under the statistical probabilistic framework, are one of the most popular techniques that have been widely applied to Information Retrieval problems. While they are famous for their well-grounded theory and good empirical performance in text retrieval, their applications in IR are often limited by their complexity and low extendability in the modeling of high-dimensional information. Recently, advances in deep learning techniques provide new opportunities for representation learning and generative models for information retrieval. In contrast to statistical models, neural models have much more flexibility because they model information and data correlation in latent spaces without explicitly relying on any prior knowledge. Previous studies on pattern recognition and natural language processing have shown that semantically meaningful representations of text, images, and many types of information can be acquired with neural models through supervised or unsupervised training. Nonetheless, the effectiveness of neural models for information retrieval is mostly unexplored. In this thesis, we study how to develop new generative models and representation learning frameworks with neural models for information retrieval. Specifically, our contributions include three main components: (1) Theoretical Analysis : We present the first theoretical analysis and adaptation of existing neural embedding models for ad-hoc retrieval tasks; (2) Design Practice : Based on our experience and knowledge, we show how to design an embedding-based neural generative model for practical information retrieval tasks such as personalized product search; And (3) Generic Framework : We further generalize our proposed neural generative framework for complicated heterogeneous information retrieval scenarios that concern text, images, knowledge entities, and their relationships. Empirical results show that the proposed neural generative framework can effectively learn information representations and construct retrieval models that outperform the state-of-the-art systems in a variety of IR tasks.

scholarly journals Multi-Class Imbalanced Graph Convolutional Network Learning

2020 ◽  
Author(s):  
Min Shi ◽  
Yufei Tang ◽  
Xingquan Zhu ◽  
David Wilson ◽  
Jianxun Liu
Keyword(s):  
State Of The Art ◽  
Representation Learning ◽  
Convolutional Network ◽  
Network Learning ◽  
Convolutional Networks ◽  
Space Experiments ◽  
Latent Distribution ◽  
Adversarial Training ◽  
Quality Of Fit

Networked data often demonstrate the Pareto principle (i.e., 80/20 rule) with skewed class distributions, where most vertices belong to a few majority classes and minority classes only contain a handful of instances. When presented with imbalanced class distributions, existing graph embedding learning tends to bias to nodes from majority classes, leaving nodes from minority classes under-trained. In this paper, we propose Dual-Regularized Graph Convolutional Networks (DR-GCN) to handle multi-class imbalanced graphs, where two types of regularization are imposed to tackle class imbalanced representation learning. To ensure that all classes are equally represented, we propose a class-conditioned adversarial training process to facilitate the separation of labeled nodes. Meanwhile, to maintain training equilibrium (i.e., retaining quality of fit across all classes), we force unlabeled nodes to follow a similar latent distribution to the labeled nodes by minimizing their difference in the embedding space. Experiments on real-world imbalanced graphs demonstrate that DR-GCN outperforms the state-of-the-art methods in node classification, graph clustering, and visualization.

Sign in / Sign up

Export Citation Format

Share Document