[This article was first published on R Works, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t.
In February, one hundred fifty-nine new packages made it to CRAN. Here are my Top 40 picks in fifteen categories: Artificial Intelligence, Computational Methods, Ecology, Genomics, Health Sciences, Mathematics, Machine Learning, Medicine, Music, Pharma, Statistics, Time Series, Utilities, Visualization, and Weather.
Artificial Intelligence
chores v0.1.0: Provides a collection of ergonomic large language model assistants designed to help you complete repetitive, hard-to-automate tasks quickly. After selecting some code, press the keyboard shortcut you’ve chosen to trigger the package app, select an assistant, and watch your chore be carried out. Users can create custom helpers just by writing some instructions in a markdown file. There are three vignettes: Getting started, Custom helpers, and Gallery.
gander v0.1.0: Provides a Copilot completion experience that knows how to talk to the objects in your R environment. ellmer chats are integrated directly into your RStudio and Positron sessions, automatically incorporating relevant context from surrounding lines of code and your global environment. See the vignette to get started.
GitAI v0.1.0: Provides functions to scan multiple Git repositories, pull content from specified files, and process it with LLMs. You can summarize the content, extract information and data, or find answers to your questions about the repositories. The output can be stored in a vector database and used for semantic search or as a part of a RAG (Retrieval Augmented Generation) prompt. See the vignette.
Computational Methods
nlpembeds v1.0.0: Provides efficient methods to compute co-occurrence matrices, point wise mutual information (PMI), and singular value decomposition (SVD), especially useful when working with huge databases in biomedical and clinical settings. Functions can be called on SQL databases, enabling the computation of co-occurrence matrices of tens of gigabytes of data, representing millions of patients over tens of years. See Hong (2021) for background and the vignette for examples.
NLPwavelet v1.0: Provides functions for Bayesian wavelet analysis using individual non-local priors as described in Sanyal & Ferreira (2017) and non-local prior mixtures as described in Sanyal (2025). See README to get started.
pnd v0.0.9: Provides functions to compute numerical derivatives including gradients, Jacobians, and Hessians through finite-difference approximations with parallel capabilities and optimal step-size selection to improve accuracy. Advanced features include computing derivatives of arbitrary order. There are three vignettes on the topics: Compatibility with numDeriv, Parallel numerical derivatives, and Step-size selection.
rmcmc v0.1.1: Provides functions to simulate Markov chains using the proposal from Livingstone and Zanella (2022) to compute MCMC estimates of expectations with respect to a target distribution on a real-valued vector space. The package also provides implementations of alternative proposal distributions, such as (Gaussian) random walk and Langevin proposals. Optionally, BridgeStan’s R interfaceBridgeStan can be used to specify the target distribution. There is an Introduction to the Barker proposal and a vignette on Adjusting the noise distribution.
sgdGMF v1.0: Implements a framework to estimate high-dimensional, generalized matrix factorization models using penalized maximum likelihood under a dispersion exponential family specification, including the stochastic gradient descent algorithm with a block-wise mini-batch strategy and an efficient adaptive learning rate schedule to stabilize convergence. All the theoretical details can be found in Castiglione et al. (2024). Also included are the alternated iterative re-weighted least squares and the quasi-Newton method with diagonal approximation of the Fisher information matrix discussed in Kidzinski et al. (2022). There are four vignettes, including introduction and residuals.
Data
acledR v0.1.0: Provides tools for working with data from ACLED (Armed Conflict Location and Event Data). Functions include simplified access to ACLED’s API, methods for keeping local versions of ACLED data up-to-date, and functions for common ACLED data transformations. See the vignette to get started.
Horsekicks v1/0/2: Provides extensions to the classical dataset Death by the kick of a horse in the Prussian Army first used by Ladislaus von Bortkeiwicz in his treatise on the Poisson distribution Das Gesetz der kleinen Zahlen. Also included are deaths by falling from a horse and by drowning. See the vignette.
wbwdi v1.0.0: Provides functions to access and analyze the World Bank’s World Development Indicators (WDI) using the corresponding API. WDI provides more than 24,000 country or region-level indicators for various contexts. See the vignette.
Ecology
rangr v1.0.6: Implements a mechanistic virtual species simulator that integrates population dynamics and dispersal to study the effects of environmental change on population growth and range shifts. Look here for background and see the vignette to get started.
Economics
godley v0.2.2: Provides tools to define, simulate, and validate stock-flow consistent (SFC) macroeconomic models by specifying governing systems of equations. Users can analyze how macroeconomic structures affect key variables, perform sensitivity analyses, introduce policy shocks, and visualize resulting economic scenarios. See Godley and Lavoie (2007), Kinsella and O’Shea (2010) for background and the vignette to get started.
Genomics
gimap v1.0.3: Helps to calculate genetic interactions in CRISPR targets by taking data from paired CRISPR screens that have been pre-processed to count tables of paired gRNA reads. Output are genetic interaction scores, the distance between the observed CRISPR score and the expected CRISPR score. See Berger et al. (2021) for background and the vignettes Quick Start, Timepoint Experiment, and Treatment Experiment.
matriz v1.0.1: Implements a workflow that provides tools to create, update, and fill literature matrices commonly used in research, specifically epidemiology and health sciences research. See README to get started.
Mathematics
flint v0.0.3: Provides an interface to FLINT, a C library for number theory which extends GNU MPFR and GNU MP with support for arithmetic in standard rings (the integers, the integers modulo n, the rational, p-adic, real, and complex numbers) as well as vectors, matrices, polynomials, and power series over rings and implements midpoint-radius interval arithmetic, in the real and complex numbers See Johansson (2017) for information on computation in arbitrary precision with rigorous propagation of errors and see the NIST Digital Library of Mathematical Functions for information on additional capabilities. Look here to get started.
Machine Learning
tall v0.1.1: Implements a general-purpose tool for analyzing textual data as a shiny application with features that include a comprehensive workflow, data cleaning, preprocessing, statistical analysis, and visualization. See the vignette.
“}
Medicine
BayesERtools v0.2.1: Provides tools that facilitate exposure-response analysis using Bayesian methods. These include a streamlined workflow for fitting types of models that are commonly used in exposure-response analysis – linear and Emax for continuous endpoints, logistic linear and logistic Emax for binary endpoints, as well as performing simulation and visualization. Look here to learn more about the workflow, and see the vignette for an overview.
SimTOST v1.0.2: Implements a Monte Carlo simulation approach to estimating sample sizes, power, and type I error rates for bio-equivalence trials that are based on the Two One-Sided Tests (TOST) procedure. Users can model complex trial scenarios, including parallel and crossover designs, intra-subject variability, and different equivalence margins. See Schuirmann (1987), Mielke et al. (2018), and Shieh (2022) for background. There are seven vignettes including Introduction and Bioequivalence Tests for Parallel Trial Designs: 2 Arms, 1 Endpoint.
Music
musicXML v1.0.1: Implements tools to facilitate data sonification and create files to share music notation in the musicXML format. Several classes are defined for basic musical objects such as note pitch, note duration, note, measure, and score. Sonification functions map data into musical attributes such as pitch, loudness, or duration. See the blog and Renard and Le Bescond (2022) for examples and the vignette to get started.
Pharma
emcAdr v1.2: Provides computational methods for detecting adverse high-order drug interactions from individual case safety reports using statistical techniques, allowing the exploration of higher-order interactions among drug cocktails. See the vignette.
SynergyLMM v1.0.1: Implements a framework for evaluating drug combination effects in preclinical in vivo studies, which provides functions to analyze longitudinal tumor growth experiments using linear mixed-effects models, perform time-dependent analyses of synergy and antagonism, evaluate model diagnostics and performance, and assess both post-hoc and a priori statistical power. See Demidenko & Miller (2019 for the calculation of drug combination synergy and Pinheiro and Bates (2000) and Gałecki & Burzykowski (2013) for information on linear mixed-effects models. The vignette offers a tutorial.
vigicaen v0.15.6: Implements a toolbox to perform the analysis of the World Health Organization (WHO) Pharmacovigilance database, VigiBase, with functions to load data, perform data management, disproportionality analysis, and descriptive statistics. Intended for pharmacovigilance routine use or studies. There are eight vignettes, including basic workflow and routine pharmacoviligance.
Psychology
cogirt v1.0.0: Provides tools to psychometrically analyze latent individual differences related to tasks, interventions, or maturational/aging effects in the context of experimental or longitudinal cognitive research using methods first described by Thomas et al. (2020). See the vignette.
Statistics
DiscreteDLM v1.0.0: Provides tools for fitting Bayesian distributed lag models (DLMs) to count or binary, longitudinal response data. Count data are fit using negative binomial regression, binary are fit using quantile regression. Lag contribution is fit via b-splines. See Dempsey and Wyse (2025) for background and README for examples.
oneinfl v1.0.1: Provides functions to estimate Estimates one-inflated positive Poisson, one-inflated zero-truncated negative binomial regression models, positive Poisson models, and zero-truncated negative binomial models along with marginal effects and their standard errors. The models and applications are described in Godwin (2024). See README for and example.
echos v1.0.3: Provides a lightweight implementation of functions and methods for fast and fully automatic time series modeling and forecasting using Echo State Networks. See the vignettes Base functions and Tidy functions.
quadVAR v0.1.2: Provides functions to estimate quadratic vector autoregression models with the strong hierarchy using the Regularization Algorithm under Marginality Principle of Hao et al. (2018) to compare the performance with linear models and construct networks with partial derivatives. See README for examples.
watcher v0.1.2: Implements an R binding for libfswatch, a file system monitoring library, that enables users to watch files or directories recursively for changes in the background. Log activity or run an R function every time a change event occurs. See the README for an example.
Visualization
jellyfisher v1.0.4: Generates interactive Jellyfish plots to visualize spatiotemporal tumor evolution by integrating sample and phylogenetic trees into a unified plot. This approach provides an intuitive way to analyze tumor heterogeneity and evolution over time and across anatomical locations. The Jellyfish plot visualization design was first introduced by Lahtinen et al. (2023). See the vignette.
xdvir v0.1-2: Provides high-level functions to render LaTeX fragments as labels and data symbols in ggplot2 plots, plus low-level functions to author, produce, and typeset LaTeX documents, and to produce, read, and render DVIfiles. See the vignette.
Weather
RFplus v1.4-0: Implements a machine learning algorithm that merges satellite and ground precipitation data using Random Forest for spatial prediction, residual modeling for bias correction, and quantile mapping for adjustment, ensuring accurate estimates across temporal scales and regions. See the vignette.
SPIChanges v0.1.0: Provides methods to improve the interpretation of the Standardized Precipitation Index under changing climate conditions. It implements the nonstationary approach of Blain et al. (2022) to detect trends in rainfall quantities and quantify the effect of such trends on the probability of a drought event occurring. There is an Introduction and a vignette Monte Carlo Experiments and Case Studies.
To leave a comment for the author, please follow the link and comment on their blog: R Works.
Analysis and Future Implications of February 2025 New CRAN Packages
Over the course of February 2025, 159 new packages made it to the Comprehensive R Archive Network (CRAN). With immense advancements in dynamic fields such as Artificial Intelligence, Genomics, Machine Learning and others, this represents another leap into a future powered by groundbreaking data-analytics too. But what does this mean for users of these packages? What longer-term implications do these hold?
Artificial Intelligence-Based Packages
Artificial Intelligence has shown significant advancements recently. The newly released packages, such as chores v0.1.0, gander v0.1.0, and GitAI v0.1.0, showcase versatile features like language model assistants, Copilot completion experience, and functions to scan Git repositories. Considering the increasing importance of automating tasks and the capabilities these packages offer, they’re expected to gain more popularity.
Actionable Advice:
Artificial Intelligence is an ever-evolving field. Stay updated with the latest advancements like large language models and more efficient programming. Learning to use new packages like chores, gander, and GitAI could help improve efficiency in automating tasks.
Computational Methods-Based Packages
New tools like nlpembeds v1.0.0, NLPwavelet v1.0, and rmcmc v0.1.1 are milestones in Computational Methods’ evolution. Such packages demonstrate the community’s focusing on computation efficiency and modeling, even with very large data sets.
Actionable Advice:
Consider updating your skills to effectively handle large volumes of data and make sense of complex data sets using packages like nlpembeds and rmcmc.
Data Packages
Twelve new data packages, including acledR v0.1.0 and Horsekicks v1/0/2, provide the community with preloaded datasets and functions to handle specific types of data efficiently. They offer potential to researchers to undertake complex studies without the hassle of preprocessing big data.
Actionable Advice:
Stay updated with the latest data packages available on CRAN to improve the efficiency of your studies and to provide a robust framework for your research.
Machine Learning Packages
A new package like tall v0.1.1 implies a user-friendly approach to analyzing textual data using machine learning. This shows a clear trend towards user-friendly, visual, and interactive tools for applied machine learning in textual data analysis.
Actionable Advice:
As a data scientist or analyst, consider deploying machine learning tools like tall in your work. It would streamline the process of extracting insights from raw textual data.
Visualization Packages
Visualization tools like jellyfisher v1.0.4 and xdvir v0.1-2 provide intuitive ways to analyze and present data, which is a crucial aspect of data analysis.
Actionable Advice:
Should you be presenting complex data sets to an audience, consider using such visualization tools to simplify consumption and interpretation.
Long-term Implications and Future Developments
CRAN’s latest package releases suggest exciting developments in fields of Artificial Intelligence, Computational Methods, Machine Learning, Data and Visualization. With the pace at which these fields are growing, professionals relying on data analysis and researchers should anticipate even more sophisticated tools and computations in the pipeline. This further indicates a clear need to keep up with understanding and ability to deploy these constantly evolving tools.
Actionable Advice:
Continually learning and applying newly released packages should be a part of your long-term strategy. This will ensure you stay ahead in the data science world, leveraging the most effective and sophisticated tools at your disposal.
[This article was first published on T. Moudiki’s Webpage – R, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t.
You can now obtain insights from your tabular data by chatting with it in techtonique.net. No plotting yet (coming soon), but you can already ask questions like:
What is the average of column A?
Show me the first 5 rows of data
Show me 5 random rows of data
What is the sum of column B?
What is the average of column A grouped by column B?
Techtonique web app, a tool designed to help you make informed, data-driven decisions using Mathematics, Statistics, Machine Learning, and Data Visualization. As of September 2024, the tool is in its beta phase (subject to crashes) and will remain completely free to use until December 24, 2024.
After registering, you will receive an email. CHECK THE SPAMS.
A few selected users will be contacted directly for feedback, but you can also send yours.
The tool is built on Techtonique and the powerful Python ecosystem. At the moment, it focuses on small datasets, with a limit of 1MB per input. Both clickable web interfaces and Application Programming Interfaces (APIs, see below) are available.
Currently, the available functionalities include:
Data visualization. Example: Which variables are correlated, and to what extent?
Probabilistic forecasting. Example: What are my projected sales for next year, including lower and upper bounds?
Machine Learning (regression or classification) for tabular datasets. Example: What is the price range of an apartment based on its age and number of rooms?
Survival analysis, analyzing time-to-event data. Example: How long might a patient live after being diagnosed with Hodgkin’s lymphoma (cancer), and how accurate is this prediction?
Reserving based on insurance claims data. Example: How much should I set aside today to cover potential accidents that may occur in the next few years?
As mentioned earlier, this tool includes both clickable web interfaces and Application Programming Interfaces (APIs).
APIs allow you to send requests from your computer to perform specific tasks on given resources. APIs are programming language-agnostic (supporting Python, R, JavaScript, etc.), relatively fast, and require no additional package installation before use. This means you can keep using your preferred programming language or legacy code/tool, as long as it can speak to the internet. What are requests and resources?
In Techtonique/APIs, resources are Statistical/Machine Learning (ML) model predictions or forecasts.
A common type of request might be to obtain sales, weather, or revenue forecasts for the next five weeks. In general, requests for tasks are short, typically involving a verb and a URL path — which leads to a response.
Below is an example. In this case, the resource we want to manage is a list of users.
– Request type (verb): GET
URL Path:http://users | Endpoint: users | API Response: Displays a list of all users
URL Path:http://users/:id | Endpoint: users/:id | API Response: Displays a specific user
– Request type (verb): POST
URL Path:http://users | Endpoint: users | API Response: Creates a new user
– Request type (verb): PUT
URL Path:http://users/:id | Endpoint: users/:id | API Response: Updates a specific user
– Request type (verb): DELETE
URL Path:http://users/:id | Endpoint: users/:id | API Response: Deletes a specific user
In Techtonique/APIs, a typical resource endpoint would be /MLmodel. Since the resources are predefined and do not need to be updated (PUT) or deleted (DELETE), every request will be a POST request to a /MLmodel, with additional parameters for the ML model.
After reading this, you can proceed to the /howtoapi page.
To leave a comment for the author, please follow the link and comment on their blog: T. Moudiki’s Webpage – R.
Techtonique.net: A New Horizon in Data Manipulation
Tabular data analysis has taken a significant turn with the recent introduction of Techtonique.net. This innovative data platform allows users to obtain crucial insights from their tabular data using an intuitive chat function. The tool, currently in its beta phase, is designed to facilitate data-driven decision-making using multiple disciplines including Mathematics, Statistics, Machine Learning, and Data Visualization.
Key Features and Functionalities
The web-based Techtonique application brings a wide range of practical functionalities to its users. Apart from getting insights from data chat, users can run R or Python code interactively in their browser. However, the application currently focuses on small datasets with a limit of 1MB per input.
Data Visualization
Data visualization is a built-in option where users can identify correlations between variables through graphical presentations.
Probabilistic Forecasting
Probabilistic forecasting is another attractive feature that allows users to predict future sales, including lower and upper bounds.
Machine Learning and Survival Analysis
Machine Learning and Survival Analysis add an extra layer to the platform by offering in-depth analysis options based on specific datasets – from predicting the price range of apartments to estimating a life insurance payout.
The Working Principles: Interactive Web Interfaces & API
Both interactive web interfaces and Application Programming Interfaces (APIs) drive the functionality in Techtonique. APIs add versatility to the platform by allowing users to send requests to perform tasks on given resources. This flexible usage is not restricted to a specific programming language and thus, supports Python, R, JavaScript, and more.
Future Developments and Long-term Implications
As Techtonique.net is still in the beta phase, one can anticipate potential updates and advancements in future releases. An evident shortcoming at this stage is the absence of a plotting function – the inclusion of which will significantly enhance the data visualization aspect of the tool. The current limit on dataset size (1MB per input) may also be addressed, offering a wider scope for large-data analysis.
In the long-term, Techtonique.net may have a profound influence on data-driven industries, creating a more convenient and efficient system for data analysis. The impact could be particularly significant within organizations dealing with large-scale data, potentially enhancing their decision-making processes and operational efficiency.
Actionable Advice
To fully utilize the potential of this cutting-edge tool, users should:
Get Familiar with the Functionality: Understand how to use the platform’s features – from running R or Python, to using the data chat capability.
Keep up with Updates: As Techtonique.net is in the beta phase, there may be frequent updates and additions – all of which could lead to improved user experience.
Provide Feedback: Be an active contributor to the system’s development by providing feedback and suggestions. Your insights could guide future enhancements and make the tool more robust and user-friendly.
Experiment with Different Technologies: The tool supports several programming languages. Therefore, use this versatile platform to enhance your skills with Python, R, and JavaScript.
[This article was first published on R Consortium, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don’t.
Haziq Jamil, the founder and organizer of the Brunei R User Group, recently spoke with the R Consortium. Haziq established the first R User Group in Brunei to promote R programming and create collaborative learning environments. Under his leadership, the group hosts monthly meetups and events to advance R skills across various sectors in Brunei. Through these efforts, Haziq aims to build a supportive and inclusive R community, encouraging both personal growth and data-driven innovation in the region.
Please share your background and involvement with the RUGS group.
My name is Haziq Jamil, and I am an Assistant Professor in Statistics at Universiti Brunei Darussalam, the leading higher education institution in Brunei. I have used R for almost ten years during my studies and on many personal and professional projects.
The Brunei R User Group was founded in February 2024, and I serve as its chair and founder. My role is to lead the group’s administration, oversee its overall direction and strategy, and ensure that its initiatives align with its mission of promoting R programming and fostering a supportive learning environment, focusing on community engagement and collaboration.
Can you share what the R community is like in Brunei?
The R community in Brunei may be small, but it is growing thanks to the efforts of the Brunei R User Group. The group organizes monthly meetups and events to promote learning and development in R programming and advance its use across various fields in Brunei. These gatherings provide opportunities to expand the community by enhancing participants’ skills, offering a platform for networking with like-minded individuals, and engaging in practical applications such as data analysis, visualization, and spatial data techniques. By creating an inclusive R community, the group aims to support individual growth in Brunei and foster collaboration on data-driven R projects. Whether for students, professionals, or hobbyists, the group strives to provide a supportive space for learning, sharing insights, and driving innovation within the local R community.
You hosted a Meetup, “R>aya with R,” in April. Can you share more about the topic? Why this topic?
The R User Group’s “R>aya with R” Meetup in Brunei was a lively event that combined Hari Raya Aidilfitri’s (Eid-ul-Fitr) festive spirit with the exploration of R programming. To engage younger audiences, we offered free boba tea as a beverage during the session. The event featured informative sessions led by expert community members, each focusing on advanced topics relevant to different fields.
One of the critical presentations was on “Survival Analysis” by Dr. Elvynna Leong. She explained statistical techniques to predict the time until an event of interest, such as guests’ arrival at a Hari Raya open house. This topic is directly related to fields that depend on time-to-event data, such as healthcare or actuarial science.
One of the highlights was Wafid Sophian’s session on “Simulation Methods for Economic Analysis.” In this presentation, Wafid demonstrated how R can be used to simulate and analyze complex economic scenarios. The topic focused on modeling outcomes and making data-driven predictions. It was chosen due to its significance in finance and business analytics.
Dr. Eden Ng presented on “Mathematical Modeling of Evolutionary Biology,” showcasing how R can be used to model biological evolutionary processes. This session visualized the intersection of R programming and biology, thus emphasizing R’s utility in research areas such as genetics and evolutionary studies.
The event covered three topics in the areas of mathematics, economics, and biology. It was open to all individuals interested in learning about the capabilities and usage of the R language, regardless of whether they were beginners or experts.
Do you recommend any techniques for planning for or during the event? (Github, Zoom, other.) Can these techniques be used to make your group more inclusive to people who cannot attend physical events in the future?
For planning and executing events like the “Analysing Spatial Data with R” event, the R>aya Meetup, and the “Introduction to R” sessions, we utilized Github. To host R scripts, datasets, and event materials. Our participants can access and review the code before and after the event. It also provides a platform for issue tracking and version control to facilitate feedback and group collaboration. It helps those who can’t attend in person to engage and contribute to our Quarto Blog. To publish event summaries, key takeaways, and additional resources on the official Brunei R User Group blog. We hope to provide a centralized location for information and help remote participants catch up.
Please share any additional details you would like to include in the blog.
As the first R user group in Brunei, we are excited to promote the adoption and growth of R across various industries. Our mission goes beyond just hosting events—we are dedicated to creating and nurturing an inclusive R community and showcasing the power of R in numerous fields.
How do I Join?
R Consortium’s R User Group and Small Conference Support Program (RUGS) provides grants to help R groups organize, share information, and support each other worldwide. We have given grants over the past four years, encompassing over 68,000 members in 33 countries. We would like to include you! Cash grants and meetup.com accounts are awarded based on the intended use of the funds and the amount of money available to distribute.
An Extended Look into Brunei’s Blossoming R Programming Community
The R programming language, a vital tool in the ever-evolving world of data analysis, is steadily gaining traction in Brunei. Spearheaded by Haziq Jamil, the founder of the Brunei R User Group, this small but growing community is continuously demonstrating the relevance of R programming across a variety of sectors, fostering an engaging learning environment that promotes personal growth, and innovation. This detailed exploration will probe into the significant implications and predictive future developments drawn from recent initiatives led by the Brunei R User Group.
Future Implications
Stronger Skills Development
With the R User Group orchestrating monthly meet-ups and hosting educational events, the R language proficiency among Brunei’s workforce can be expected to witness a significant surge. Companies in a wide array of sectors, including but not limited to healthcare, actuarial science, finance, and business analytics, could therefore be leveraging R programming for more robust data examination. This may also improve the competitive advantage of local businesses in the long run.
Enhanced Collaboration
The inclusive community environment that nurtures collaboration ensures the likelihood of increased data-driven projects in Brunei’s future. By collectively harnessing the power of R, this emerging assembly could spur great advancements within and beyond the realm of data science.
Elevated Innovation
Through cultivating individual growth and enhancing R capabilities, the likelihood of innovative solutions emanating from the Brunei R User Group is substantial. Circulating these innovations across various industries could incite notable changes in how local companies operate and make decisions based on data.
Possible Future Developments
Expansion of the R-Adopting Community
As the efforts in promoting R programming continue, the Brunei R User Group’s size is likely to experience growth. With the findings and advancements shared in these meetings becoming more impactful, a growing number of individuals and companies will start tapping into the potential of R programming for their data analysis needs.
Increased Global Support
Gaining more attention at a global level, significant funds and resources could come the way of the Brunei R User Group. The R Consortium’s R User Group and Small Conference Support Program (RUGS) has already proven crucial in propelling R programming across the globe, providing substantial backing for user groups in 33 countries. It’s plausible that Brunei’s R User Group will see increased support over the years.
Actionable Advice
Embrace R Programming: For businesses and institutions, it would be beneficial to consider R programming as a viable tool in data analysis. The resulting insights and predictions derived from sophisticated R-models may offer significant competitive advantages.
Support Local Talent: It’s crucial for the country’s sectors to support and invest in local talent. Encouraging participation in the Brunei R User Group will foster a vibrant pool of competent professionals most adept at harnessing the power of R.
Seek Global Collaboration: The R User Group should continuously seek to establish international networks for an exchange of invaluable insights and methodologies. Such collaborations could provide global solutions to local problems and, in turn, elevate the group’s status on an international scale.
“The power of R programming lies in its ability to uncover robust insights from complex data. As such, the Brunei R User Group continues to foster an inclusive environment that encourages learning, collaboration, and innovation – the key components of a prosperous future.”
arXiv:2405.14959v1 Announce Type: new Abstract: Event cameras offer promising advantages such as high dynamic range and low latency, making them well-suited for challenging lighting conditions and fast-moving scenarios. However, reconstructing 3D scenes from raw event streams is difficult because event data is sparse and does not carry absolute color information. To release its potential in 3D reconstruction, we propose the first event-based generalizable 3D reconstruction framework, called EvGGS, which reconstructs scenes as 3D Gaussians from only event input in a feedforward manner and can generalize to unseen cases without any retraining. This framework includes a depth estimation module, an intensity reconstruction module, and a Gaussian regression module. These submodules connect in a cascading manner, and we collaboratively train them with a designed joint loss to make them mutually promote. To facilitate related studies, we build a novel event-based 3D dataset with various material objects and calibrated labels of grayscale images, depth maps, camera poses, and silhouettes. Experiments show models that have jointly trained significantly outperform those trained individually. Our approach performs better than all baselines in reconstruction quality, and depth/intensity predictions with satisfactory rendering speed.
This article introduces a groundbreaking framework called EvGGS, which aims to overcome the challenges of reconstructing 3D scenes from raw event streams captured by event cameras. Event cameras have unique advantages such as high dynamic range and low latency, making them ideal for challenging lighting conditions and fast-moving scenarios. However, the sparse and colorless nature of event data makes 3D reconstruction difficult. EvGGS is the first event-based generalizable 3D reconstruction framework that can reconstruct scenes as 3D Gaussians solely from event input in a feedforward manner. What sets EvGGS apart is its ability to generalize to unseen cases without the need for retraining. The framework consists of a depth estimation module, an intensity reconstruction module, and a Gaussian regression module, all of which are jointly trained with a designed joint loss to enhance their performance. To support further research in this field, the authors have also created a novel event-based 3D dataset with various material objects and calibrated labels. Experimental results demonstrate that the jointly trained models outperform individually trained ones, achieving superior reconstruction quality and accurate depth/intensity predictions at a satisfactory rendering speed.
An Innovative Approach to Event-Based 3D Scene Reconstruction
The field of 3D reconstruction has seen rapid advancements in recent years, allowing us to capture and represent the world in three dimensions. Traditional methods heavily rely on RGB images and depth sensors, which can be limited by challenging lighting conditions and fast-moving scenarios. However, a new technology called event cameras has emerged, offering promising advantages such as high dynamic range and low latency, which make them well-suited for these challenging scenarios.
Event cameras capture the changes in the scene asynchronously, producing a continuous stream of events. Each event consists of the pixel coordinates, the timestamp, and the sign indicating whether the change was a decrease or increase in intensity. However, reconstructing 3D scenes from this sparse event data is a challenging task, as it does not carry absolute color information like traditional RGB images.
To unlock the true potential of event cameras in 3D reconstruction, a team of researchers has proposed an innovative framework called EvGGS (Event-based Generalizable 3D Gaussian Reconstruction). This framework is the first of its kind to reconstruct scenes as 3D Gaussians solely from event input, in a feedforward manner, and without requiring retraining for unseen cases.
EvGGS consists of three key submodules: a depth estimation module, an intensity reconstruction module, and a Gaussian regression module. These submodules work in a cascading manner, where the output of one submodule feeds into the next. To ensure the collaboration and mutual promotion of these submodules, they are jointly trained using a specially designed loss function.
In order to facilitate further research in this domain, the researchers have also built a novel event-based 3D dataset. This dataset contains various material objects along with calibrated labels of grayscale images, depth maps, camera poses, and silhouettes. This dataset will serve as a valuable resource for other researchers interested in exploring event-based 3D reconstruction techniques.
The experiments conducted by the researchers demonstrate the effectiveness of their approach. The jointly trained models significantly outperform those trained individually, both in terms of reconstruction quality and depth/intensity predictions. Furthermore, the proposed framework achieves satisfactory rendering speed.
EvGGS opens up new possibilities for event-based 3D scene reconstruction. By leveraging the unique advantages of event cameras, this framework enables accurate and reliable reconstructions in challenging lighting conditions and fast-moving scenarios. The ability to generalize to unseen cases without the need for retraining is a groundbreaking achievement that paves the way for real-world applications of event-based 3D reconstruction.
“EvGGS represents a paradigm shift in the field of event-based 3D reconstruction. It combines the power of event cameras with the versatility of 3D Gaussians, providing a robust and efficient solution for capturing the dynamic world in three dimensions. This research marks a significant step towards bridging the gap between traditional RGB-based approaches and the rapidly evolving event-based paradigm.”
In conclusion, EvGGS is a pioneering framework that pushes the boundaries of event-based 3D scene reconstruction. Its ability to reconstruct scenes as 3D Gaussians solely from event input, combined with its generalizability and joint training approach, make it a game-changer in this field. With the built-in dataset and promising experimental results, EvGGS sets a new standard for event-based 3D reconstruction and opens up exciting avenues for future research and real-world applications.
The paper introduces a novel framework called EvGGS, which aims to address the challenges of reconstructing 3D scenes from raw event streams captured by event cameras. Event cameras, known for their high dynamic range and low latency, are particularly well-suited for challenging lighting conditions and fast-moving scenarios. However, the sparsity of event data and the absence of absolute color information make 3D reconstruction a difficult task.
EvGGS is the first event-based generalizable 3D reconstruction framework that operates in a feedforward manner, meaning it can reconstruct scenes as 3D Gaussians using only event input. What sets EvGGS apart is its ability to generalize to unseen cases without the need for retraining. This is a significant advancement in the field, as most existing methods require retraining or fine-tuning when faced with new scenarios.
The framework consists of three interconnected submodules: a depth estimation module, an intensity reconstruction module, and a Gaussian regression module. These modules are trained collaboratively with a joint loss, which encourages them to mutually promote each other’s performance. By cascading these submodules, EvGGS can reconstruct 3D scenes from event data in an efficient and accurate manner.
To support further research and evaluation, the authors have also created a new event-based 3D dataset. This dataset includes various material objects and provides calibrated labels for grayscale images, depth maps, camera poses, and silhouettes. The availability of this dataset is expected to facilitate future studies in the field of event-based 3D reconstruction.
The experiments conducted by the authors demonstrate the effectiveness of the EvGGS framework. Models trained jointly outperform those trained individually, indicating the benefits of the collaborative training approach. EvGGS shows superior reconstruction quality compared to baseline methods and achieves satisfactory rendering speed for depth and intensity predictions.
Overall, the introduction of EvGGS represents a significant step forward in event-based 3D reconstruction. Its ability to generalize to unseen cases without retraining, combined with its improved reconstruction quality and rendering speed, make it a promising framework for various applications, including robotics, augmented reality, and autonomous vehicles. As this research gains traction, it will be interesting to see how the EvGGS framework evolves and potentially integrates with other emerging technologies in the field. Read the original article
We introduce an online mathematical framework for survival analysis, allowing real time adaptation to dynamic environments and censored data. This framework enables the estimation of event time…
In the fast-paced world we live in, it is crucial to have tools that can adapt to changing environments and handle complex data. In this article, we present an innovative online mathematical framework for survival analysis that does just that. Our framework not only allows for real-time adaptation to dynamic environments but also handles censored data, providing accurate estimations of event time. With this cutting-edge tool, researchers and analysts can now navigate the complexities of survival analysis with ease, unlocking valuable insights in various fields such as healthcare, finance, and social sciences.
Survival Analysis in a Dynamic Environment: A New Mathematical Framework
Survival analysis has long been an essential tool in various fields such as medicine, engineering, and economics. It involves the study of time-to-event data, where events can be anything from the occurrence of a disease to the failure of a mechanical component. Traditionally, survival analysis has focused on analyzing static environments with complete data. However, in today’s fast-paced and ever-changing world, it is crucial to have a framework that can adapt to dynamic environments and handle censored data.
The Challenges of Dynamic Environments
In many real-world scenarios, the factors affecting event times can change over time. For example, in healthcare, the effectiveness of a treatment can vary over different periods as new drugs or therapies are introduced. Similarly, in engineering, the failure rate of a component may change as it ages or when external conditions vary. Traditional survival analysis methods often fail to account for these dynamic factors, leading to inaccurate estimations and predictions.
Censored data poses another challenge in survival analysis. Censoring occurs when the event of interest has not yet occurred for some individuals by the end of the study or observation period. Handling censored data requires sophisticated methods that can properly incorporate this partial information into the analysis.
An Online Mathematical Framework
Addressing the limitations of existing approaches, we propose an online mathematical framework for survival analysis. This framework allows real-time adaptation to dynamic environments and handles censored data in a robust manner. Our method combines elements from machine learning, statistical modeling, and optimization techniques to provide accurate estimations and predictions even in rapidly changing scenarios.
The core idea behind our framework is to continuously update and refine the survival models as new data becomes available. By leveraging online learning algorithms, we can adapt the models to changing conditions and make adjustments to the estimated survival probabilities. This dynamic approach ensures that the analysis stays relevant and reliable in real-time.
Innovative Solutions and Ideas
Our framework offers several innovative solutions to common challenges in survival analysis:
Adaptive Survival Modeling: By using online learning algorithms, our framework can adapt the survival models to changing environments. This allows for more accurate estimations of event times, especially when the underlying factors are dynamic.
Handling Censored Data: Our framework incorporates censored data by utilizing advanced statistical techniques. It considers the partial information provided by censored observations, improving the accuracy of the analysis.
Real-time Predictions: With its ability to adapt to dynamic environments, our framework enables real-time predictions of event times. This is particularly valuable in situations where timely decisions need to be made, such as healthcare interventions or preventative maintenance in engineering.
Flexible Implementation: Our framework can be implemented in various domains and can handle different types of event data. It provides a flexible solution that can be customized to specific needs and requirements.
Survival analysis in a dynamic environment requires an innovative and adaptive approach. Our online mathematical framework offers a robust solution for handling dynamic factors and censored data. By continuously updating the models and incorporating new information in real time, our framework provides accurate estimations and predictions. This opens up new possibilities for decision-making in fields such as healthcare, engineering, and beyond.
and survival probabilities in complex scenarios, such as medical research and actuarial science, where time-to-event data is commonly encountered. Survival analysis, also known as time-to-event analysis, is a statistical technique used to analyze the time it takes for an event of interest to occur, such as death, failure of a system, or occurrence of a disease.
The development of an online mathematical framework for survival analysis is a significant advancement in this field. Traditionally, survival analysis has been performed using static models that assume the data is fixed and does not change over time. However, in many real-world applications, the data is dynamic and subject to censoring, where the event of interest has not yet occurred for some subjects at the time of analysis.
By introducing an online framework, researchers and practitioners can now adapt their models and estimates in real time as new data becomes available. This is particularly valuable in situations where the environment is constantly changing, such as in clinical trials or monitoring the progression of diseases.
One key advantage of this framework is its ability to handle censored data. Censoring occurs when the event of interest has not occurred for some subjects within the study period or follow-up time. Traditional methods often treat censored observations as missing data or exclude them from the analysis, leading to biased results. The online framework, however, incorporates these censored observations and provides more accurate estimates of survival probabilities and event times.
Moreover, the online nature of this framework allows for continuous updating of estimates as new data points are collected. This feature is particularly useful in scenarios where data collection is ongoing or when there are delays in obtaining complete information. Researchers can now make more informed and timely decisions based on the most up-to-date information available.
Looking ahead, there are several potential avenues for further development and application of this online mathematical framework for survival analysis. One direction could be to incorporate machine learning techniques to enhance predictive capabilities and identify patterns in the data that may not be captured by traditional parametric models. Additionally, the framework could be extended to handle competing risks, where multiple events of interest may occur, and the occurrence of one event may affect the probability of others.
Furthermore, the implementation of this framework in real-world settings, such as healthcare systems or insurance industries, could provide valuable insights into predicting patient outcomes, optimizing treatment strategies, or assessing risk profiles. By continuously updating survival estimates based on newly collected data, healthcare providers and insurers can make more accurate assessments of individual patient risk and tailor interventions accordingly.
In conclusion, the introduction of an online mathematical framework for survival analysis is a significant advancement in the field. Its ability to adapt to dynamic environments and handle censored data opens up new possibilities for accurate estimation of event times and survival probabilities. This framework has the potential to revolutionize various domains, including medical research, healthcare, and actuarial science, by enabling real-time decision-making and personalized interventions based on the most up-to-date information available. Read the original article
“}
oneinfl v1.0.1: Provides functions to estimate Estimates one-inflated positive Poisson, one-inflated zero-truncated negative binomial regression models, positive Poisson models, and zero-truncated negative binomial models along with marginal effects and their standard errors. The models and applications are described in Godwin (2024). See README for and example.
quadVAR v0.1.2: Provides functions to estimate quadratic vector autoregression models with the strong hierarchy using the Regularization Algorithm under Marginality Principle of Hao et al. (2018) to compare the performance with linear models and construct networks with partial derivatives. See README for examples.
