up

content/code/feature_extraction/hocca/content.md

@@ -1,6 +1,7 @@
 ---
 title: "Spatio-Chromatic Adaptation via Higher-Order Canonical Correlation Analysis of Natural Images"
 abstract: "Independent component and canonical correlation analysis are twogeneral-purpose statistical methods with wide applicability. Inneuroscience, independent component analysis of chromatic naturalimages explains the spatio-chromatic structure of primary corticalreceptive fields in terms of properties of the visual environment.Canonical correlation analysis explains similarly chromatic adaptationto different illuminations. But, as we show in this paper, neither ofthe two methods generalizes well to explain both spatio-chromaticprocessing and adaptation at the same time. We propose a statisticalmethod which combines the desirable properties of independent componentand canonical correlation analysis: It finds independent components ineach data set which, across the two data sets, are related to eachother via linear or higher-order correlations. The new method is aswidely applicable as canonical correlation analysis, and also to morethan two data sets. We call it higher-order canonical correlationanalysis. When applied to chromatic natural images, we found that itprovides a single (unified) statistical framework which accounts forboth spatio-chromatic processing and adaptation. Filters withspatio-chromatic tuning properties as in the primary visual cortexemerged and corresponding-colors psychophysics was reproducedreasonably well. We used the new method to make a theory-driventestable prediction on how the neural response to colored patternsshould change when the illumination changes. We predict shifts in theresponses which are comparable to the shifts reported for chromaticcontrast habituation."
+
 referencias:
  - nombre: "Spatio-Chromatic Adaptation via Higher-Order Canonical Correlation Analysis of Natural Images"
  autores: "M. U. Gutmann, V. Laparra, A. Hyvärinen, J. Malo"

content/code/feature_extraction/spca/content.md

@@ -24,14 +24,15 @@ imagenes:
  - ruta: "spca_im_coding2.webp"
  titulo: "Image Coding Example 2"
  descripcion: "Further results of image coding with different metrics."
+
 referencias:
  - nombre: "V1 non-linear properties emerge from local-to-global non-linear ICA"
  autores: "J. Malo, J. Gutiérrez"
  publicacion: "Network: Comp. Neural Systems. 17(1): 85-102 (2006)"
  url: "https://huggingface.co/datasets/isp-uv-es/Web_site_legacy/resolve/main/code/soft_feature/V1_from_non_linear_ICA.pdf"
  - nombre: "Visual Aftereffects and Sensory Nonlinearities from a Single Statistical Framework"
  autores: "V. Laparra, J. Malo"
- publicacion: "Frontiers in Human Neuroscience. [Special Issue on Perceptual Illusions](http://journal.frontiersin.org/researchtopic/the-future-of-perceptual-illusions-from-phenomenology-to-neuroscience-2381) 2015. [A guide to the full supplementary material (description of the code, data, experiments and results)](https://ipl-uv.github.io/old_pages/data/after_effects/index.html)."
+ publicacion: "Frontiers in Human Neuroscience. [Special Issue on Perceptual Illusions](http://journal.frontiersin.org/researchtopic/the-future-of-perceptual-illusions-from-phenomenology-to-neuroscience-2381) 2015. [A guide to the full supplementary material (description of the code, data, experiments and results)](../../../vision_and_color/aftereffects/content/)."
  url: "https://huggingface.co/datasets/isp-uv-es/Web_site_legacy/resolve/main/code/soft_feature/frontiers_laparra_malo_Accepted_15.pdf"
  - nombre: "Nonlinearities and Adaptation of Color Vision from Sequential Principal Curves Analysis"
  autores: "V. Laparra, S. Jiménez, G. Camps, J. Malo"

content/code/vision_and_color/aftereffects/content.md

@@ -67,7 +67,7 @@ abstract: |
 
  Rather than using specific functional forms for the adaptation, in this work we derive the behavior from a recently proposed unsupervised non-parametric learning technique: The Sequential Principal Curves Analysis (SPCA) [Laparra et al. 12]. SPCA effectively performs the multidimensional equalization previously suggested, but not implemented, in the statistically inspired literature addressing aftereffects [Clifford00, Clifford02]. We argue that unsupervised learning is the appropriate way to focus on the principle behind the aftereffects: as we assume no parametric form, it is more clear that the behavior emerges from the specific optimization strategies and not from an a priori response model. Moreover, unlike other unsupervised learning techniques, SPCA is more suited to answer the goal question because it can be easily tuned to different principles such as information maximization (as non-linear ICA) and also error minimization in limited resolution scenarios (as in optimal Vector Quantization). See Section 4 in the Paper to see the equalization capabilities of SPCA on visual textures. The illustrations below show how equalization leads to attenuations and shifts in the responses that induce the aftereffects.
 
- Our results ([reproducible using this SPCA implementation](predictions-using-sequential-principal-curves-analysis) in these (numerical experiments)[#code] over these [natural video and calibrated image databases](#code)) show that what appear to be dysfunctional illusions are actually by-products of optimality principles such as maximum information extraction or error minimization in the representation.
+ Our results ([reproducible using this SPCA implementation](../../../feature_extraction/spca/content/) in these (numerical experiments)[#code] over these [natural video and calibrated image databases](#code)) show that what appear to be dysfunctional illusions are actually by-products of optimality principles such as maximum information extraction or error minimization in the representation.
 
  # Predictions using Sequential Principal Curves Analysis
 

content/news/2016/sedal-grant/_index.md

@@ -1,7 +1,6 @@
 ---
 title: "SEDAL Grant"
 date: 2016-01-01
-draft: false
 externalLink: "./2016/sedal-grant/sedal/index.html"
 ---
 

content/news/2016/sedal-grant/main_objetives.md

@@ -1,24 +1,30 @@
+---
+title: "SEDAL: Statistical Learning for Earth Observation Data Analysis"
+type: "news"
+layout: "single"
+---
+
 # Objectives
 
 **Improve prediction models by adaptation to Earth Observation data characteristics.** We will rely on the framework of kernel learning, which has emerged as the most appropriate framework for remote sensing data analysis in the last decade. The new retrieval models will be adapted to the particular signal characteristics, such as unevenly sampled time series and missing data, non-Gaussianity, presence of heteroscedastic and non-stationary processes, and non-i.i.d. (spatial and temporal) relations. Models based on kernels and GPs will allow us to advance in uncertainty quantification using predictive variances under biophysical constraints. Advances in sparse, reduced-rank and divide-and-conquer schemes will address the computational cost problem. The proposed kernel framework aims to improve results in terms of accuracy, reduced uncertainty, consistency of the estimations and computational efficiency.
 
+<br>
+
 **Discover knowledge and causal relations in Earth observation data.** We will investigate graphical causal models and regression-based causal schemes applied to large heterogeneous EO data streams. This will require improved measures of (conditional) independence, designing experiments in controlled situations and using high-quality data. Learning the hierarchy of the relations between variables and related covariates, as well as their causal relations, may in turn allow the discovery of hidden essential variables, drivers and confounders. Moving from correlation to dependence and then to causation concepts is fundamental to advance the field of Earth Observation and the science of climate change.
 
 # Research
 
-SEDAL aims at contributing novel machine learning algorithms along these lines:
-
-- Advanced remote sensing data and EO time series processing and statistical characterization
-- Advanced regression methods, involving kernel methods, Gaussian processes, random forests, and deep nets
-- Efficient large-scale model implementations
-- Uncertainty quantification and propagation
-- Physically-based models, emulation of RTMs, and design of physically-meaningful priors in machine learning regression
-- Knowledge discovery and structure learning from empirical EO data
-- (Conditional) Dependence estimation of EO variables and observations
-- Graphical models, structure learning, Bayesian networks and causal inference from empirical EO data
-
-The target EO applications are:
+- SEDAL aims at contributing novel machine learning algorithms along these lines:
+ - Advanced remote sensing data and EO time series processing and statistical characterization
+ - Advanced regression methods, involving kernel methods, Gaussian processes, random forests, and deep nets
+ - Efficient large-scale model implementations
+ - Uncertainty quantification and propagation
+ - Physically-based models, emulation of RTMs, and design of physically-meaningful priors in machine learning regression
+ - Knowledge discovery and structure learning from empirical EO data
+ - (Conditional) Dependence estimation of EO variables and observations
+ - Graphical models, structure learning, Bayesian networks and causal inference from empirical EO data
 
-- Improved retrieval (regression) algorithms at local, regional, and global planetary scales
-- Structure inference and relevance determination of essential climate variables and observations
-- Climate change detection, anomalies, extremes, and causal inference attribution
+- The target EO applications are:
+ - Improved retrieval (regression) algorithms at local, regional, and global planetary scales
+ - Structure inference and relevance determination of essential climate variables and observations
+ - Climate change detection, anomalies, extremes, and causal inference attribution

content/news/2016/sedal-grant/motivation.md → content/news/2016/sedal-grant/motivation/_index.md

@@ -1,30 +1,11 @@
-# Motivation
+---
+title: "Motivation"
+type: "research"
+layout: "single"
+---
 
 Despite the many successful results and developments, there are still strong limitations for the general adoption of machine learning algorithms for predicting and understanding EO data. Machine learning and signal processing have advanced enormously in the last decade (both at theoretical and applied levels) but have not moved forward the field of EO data analysis to its full potential.
 
 The current statistical treatment of biophysical parameters is strongly limited by the quantity and quality of EO data, as well as by the abuse of standard off-the-shelf methods, which, in general, are not well-adapted to the particular characteristics of EO data. Specifically, current regression models used for EO applications are still deficient because they rely on limited amounts of meteorological and remote sensing data, do not observe the particular data characteristics, and often make strong assumptions of linearity, homoscedasticity, or Gaussianity. These limitations translate into certain risks of overfitting and unreasonably large uncertainties for the predictions, suggesting a lack of explanatory variables and deficiencies in model specification. Graphical models have been seldom used in EO data analysis. The few works restrict to local studies, use limited amounts of data and explanatory variables, consider remote sensing input features only, apply standard structure learning algorithms driven by univariate (often unconditioned) dependence estimates, and do not extract causal relations or identify new drivers in the problem.
 
-We advocate that machine learning algorithms for EO applications need to be guided both by data and by prior physical knowledge. This combination is the way to restrict the family of possible solutions and thus obtain nonparametric flexible models that respect the physical rules governing the Earth climate system. We are equally concerned about the ‘black-box’ criticism of statistical learning algorithms, for which we aim to design self-explanatory models and take a leap towards the relevant concept of causal inference from empirical EO data.
-
-# Related Projects
-
-<br>
-
-## Cloud Detection in the Cloud
-- **Google Earth Engine Research Award, L. Gomez-Chova**
-- 01/16 - 12/17
-- [Cloud detection in the cloud](/old_pages/other/cloud_detection.html)
-
-## LIFE-VISION: Learning Image Features to Encode Visual Information
-- **Spanish Ministry of Economy and Competitiveness, 2012. TIN2012-38102-C03-01**
-- 01/13 - 12/15
-- [LIFE-VISION](http://lifevisionproject.wordpress.com/)
-
-## GEOLEARN
-- **Spanish Ministry of Economy and Competitiveness**
-- 2016
-- [GEOLEARN](/old_pages/other/motivation_sd.html)
-
-## ESA CCI Soil Moisture
-- **European Space Agency**
-- [ESA CCI Soil Moisture](http://esa-soilmoisture-cci.org/)
+We advocate that machine learning algorithms for EO applications need to be guided both by data and by prior physical knowledge. This combination is the way to restrict the family of possible solutions and thus obtain nonparametric flexible models that respect the physical rules governing the Earth climate system. We are equally concerned about the ‘black-box’ criticism of statistical learning algorithms, for which we aim to design self-explanatory models and take a leap towards the relevant concept of causal inference from empirical EO data.

content/news/2016/sedal-grant/motivation/cci-sm2.md

@@ -0,0 +1,12 @@
+---
+title: 'ESA Climate Change Initiative Phase II Soil Moisture (CCI SM 2 Project)'
+logo: 'esa.webp'
+pi: 'M. Piles (Visiting Scientist), Wouter Dorigo (TU Wien, PI)'
+uvpi: ''
+years: '2017'
+website: 'http://www.esa-soilmoisture-cci.org/'
+funding_source: 'ESA'
+role: ''
+project_type: ''
+partners: []
+---

content/news/2016/sedal-grant/motivation/cloud-detection-in-the-cloud.md

@@ -0,0 +1,14 @@
+---
+title: 'Cloud Detection in the Cloud'
+logo: 'google.webp'
+pi: ''
+uvpi: 'L. Gomez-Chova'
+years: '2016-2017'
+website: 'https://isp.uv.es/projects/cdc/GEE_cloud_detection_results.html'
+funding_source: 'Google Earth Engine Research Award'
+role: ''
+project_type: ''
+partners: []
+type: "news"
+layout: "single_projects"
+---

content/news/2016/sedal-grant/motivation/esa-climate-change-initiative.md

@@ -0,0 +1,12 @@
+---
+title: 'ESA Climate Change Initiative (CCI) Phase 1: Essential Climate Variable (ECV) Cloud'
+logo: 'esa.webp'
+pi: ''
+uvpi: ''
+years: '2010-2013'
+website: 'http://www.esa-cci.org/'
+funding_source: 'ESA'
+role: ''
+project_type: ''
+partners: []
+---

content/news/2016/sedal-grant/motivation/geolearn.md

@@ -0,0 +1,12 @@
+---
+title: 'GEOLEARN: Advances in Machine Learning for Large Scale Remote Sensing Data Processing'
+logo: 'mineco.webp'
+pi: ''
+uvpi: 'Jordi Munoz-Mari'
+years: '2015-2018'
+website: 'http://harpo.uv.es/wiki/geolearn:start'
+funding_source: 'Spanish Ministry of Economy and Competitiveness'
+role: ''
+project_type: ''
+partners: []
+---

content/news/2016/sedal-grant/motivation/life-vision.md

@@ -0,0 +1,12 @@
+---
+title: 'LIFE-VISION: Learning Image Features to Encode Visual Information'
+logo: 'mineco.webp'
+pi: ''
+uvpi: ''
+years: '2012-2015'
+website: 'http://lifevisionproject.wordpress.com/'
+funding_source: 'Spanish Ministry of Economy and Competitiveness'
+role: ''
+project_type: ''
+partners: []
+---

content/news/2016/sedal-grant/project_structure.md

@@ -1,8 +1,12 @@
-# Methodology
+---
+title: "Methodology"
+type: "news"
+layout: "single"
+---
 
 Activities are organized in three major tasks: two theoretical tasks guided by an application-oriented one dealing with relevant EO problems.
 
-## Workpackage 1: Improving Statistical Regression Models
+# Workpackage 1: Improving Statistical Regression Models
 
 We will develop new kernel regression models to cope with the shortcomings identified before, namely: improve model’s accuracy by encoding prior knowledge, quantify the uncertainty of the estimations, attain self-explanatory models, and alleviate the computational cost. We will develop ways to encode prior knowledge about the problem by design of kernels and neural structures able to:
 
@@ -14,7 +18,7 @@ We will develop new kernel regression models to cope with the shortcomings ident
 6. Deploy efficient (sparse and divide-and-conquer) kernel regression models.
 7. Discover knowledge in kernel models.
 
-## Workpackage 2: Learning Graphical Models and Causal Inference
+# Workpackage 2: Learning Graphical Models and Causal Inference
 
 We will exploit results and algorithms of the previous task in order to develop methods that can learn nonlinear data dependencies and possibly infer causal relations. We will propose:
 
@@ -25,20 +29,20 @@ We will exploit results and algorithms of the previous task in order to develop
 
 Models and inferred structures will be tested through pure non-interventional settings, as well as intervention analyses in controlled situations, that might reveal the presence of hidden causal variables and relationships, and by quantifying the impact of prior (physical) knowledge.
 
-## Workpackage 3: Case Study - From Local to Global Scales in EO Variable Learning
+# Workpackage 3: Case Study - From Local to Global Scales in EO Variable Learning
 
 We will focus on the relevant applications of:
 
 1. Learning statistical predictive models for key biophysical variables.
 2. Extracting knowledge from the models and the nonlinear hierarchical data representations.
 3. Inferring causal variable relations from empirical data, both at local and global scales.
 
-### Local Scale
+## Local Scale
 
 - Modeling biophysical parameters at local scale, primarily focusing on chlorophyll content, fluorescence, biomass, LAI, and fAPAR.
 - The study and quantification of uncertainty, inclusion of prior physical knowledge to constrain model’s flexibility, and the analysis of dependence/causal relations between variables will be the main scientific questions to be addressed.
 
-### Global Scale
+## Global Scale
 
 - Generate global flux products derived from upscaling FLUXNET eddy covariance observations using an array of remote sensing data.
 - We will evaluate the developed regression algorithms, the relative relevance of explanatory variables, and will learn graph dependencies between remote sensing variables and carbon (e.g., total ecosystem respiration, net ecosystem exchange), energy (e.g., latent heat and heat radiation), and water (e.g., evapotranspiration) fluxes.

content/news/2016/sedal-grant/proporsal.md

@@ -1,28 +1,31 @@
-# SEDAL Project
+---
+title: "SEDAL Project"
+type: "news"
+layout: "single"
+---
 
-<br>
-
-## Proposals
+# Proposals
 
 - **B1 Proposal**
 - **B2 Proposal**
 
-## Interview Slides
+# Interview Slides
 
-<br>
 
-### Reporting
+## Reporting
 
 - **Continuous Reporting:** 01/09/2015 - 28/02/2017
 - **Mid-term Report:** 01/09/2015 - 28/02/2018
 
-### Outreach Presentations
+## Outreach Presentations
 
 - **Advanced Applications in AI (AAA)**
 - **Algorithms and Analysis (AAA)**
 - **Applied Analytics for Agriculture (AAA)**
 - **Atmospheric and Aerial Analysis (AAA)**
 
+<br>
+
 The SEDAL project is an interdisciplinary effort to develop novel statistical learning methods to analyze Earth Observation (EO) satellite data. The project focuses on improving prediction models, discovering knowledge and causal relations in EO data, and contributing to various remote sensing applications.
 
 Through the development of kernel learning frameworks and graphical models, SEDAL aims to address current limitations in EO data analysis. The project's methodologies involve enhancing statistical regression models, learning graphical models and causal inference, and conducting case studies from local to global scales.