computer modelling in problem solving

Help | Advanced Search

Computer Science > Computation and Language

Title: geocoder: solving geometry problems by generating modular code through vision-language models.

Abstract: Geometry problem-solving demands advanced reasoning abilities to process multimodal inputs and employ mathematical knowledge effectively. Vision-language models (VLMs) have made significant progress in various multimodal tasks. Yet, they still struggle with geometry problems and are significantly limited by their inability to perform mathematical operations not seen during pre-training, such as calculating the cosine of an arbitrary angle, and by difficulties in correctly applying relevant geometry formulas. To overcome these challenges, we present GeoCoder, which leverages modular code-finetuning to generate and execute code using a predefined geometry function library. By executing the code, we achieve accurate and deterministic calculations, contrasting the stochastic nature of autoregressive token prediction, while the function library minimizes errors in formula usage. We also propose a multimodal retrieval-augmented variant of GeoCoder, named RAG-GeoCoder, which incorporates a non-parametric memory module for retrieving functions from the geometry library, thereby reducing reliance on parametric memory. Our modular code-finetuning approach enhances the geometric reasoning capabilities of VLMs, yielding an average improvement of over 16% across various question complexities on the GeomVerse dataset compared to other finetuning methods.

Submission history

Access paper:.

• HTML (experimental)
• Other Formats

References & Citations

• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

WATCH THE PROGRAM DEMO VIDEO

By submitting your information, you are agreeing to receive periodic information about online programs from MIT related to the content of this course.

Machine Learning, Modeling, and Simulation: Engineering Problem-Solving in the Age of AI

Demystify machine learning through computational engineering principles and applications in this two-course program from MIT xPRO

Submit your information to discover what makes this machine learning program different and how you'll learn with MIT xPRO.

A HANDS-ON APPROACH TO ENGINEERING PROBLEM-SOLVING

The advent of big data, cloud computing, and machine learning are revolutionizing how many professionals approach their work. These technologies offer exciting new ways for engineers to tackle real-world challenges. But with little exposure to these new computational methods, engineers lacking data science or experience in modern computational methods might feel left behind.

This two-course online certificate program brings a hands-on approach to understanding the computational tools used in modern engineering problem-solving. 

Leveraging the rich experience of the faculty at the MIT Center for Computational Science and Engineering (CCSE), this program connects your science and engineering skills to the principles of machine learning and data science. With an emphasis on the application of these methods, you will put these new skills into practice in real time.

February 3, 2025

5 weeks per course.

AFTER THIS PROGRAM, YOU WILL:

Learn how to simulate complex physical processes in your work using discretization methods and numerical algorithms.

Assess and respond to cost-accuracy tradeoffs in simulation and optimization, and make decisions about how to deploy computational resources.

Understand optimization techniques and their fundamental role in machine learning.

Practice real-world forecasting and risk assessment using probabilistic methods.

Recognize the limitations of machine learning and what MIT researchers are doing to resolve them.

Learn about current research in machine learning at the MIT CCSE and how it might impact your work in the future.

COURSES IN THIS PROGRAM

Machine Learning, Modeling, and Simulation Principles

Course 1 of 2 in the Machine Learning, Modeling, and Simulation online program

View Weekly Schedule

Applying Machine Learning to Engineering and Science

Course 2 of 2 in the Machine Learning, Modeling, and Simulation online program

View WEEKLY SCHEDULE

The MIT xPRO Learning Experience

We bring together an innovative pedagogy paired with world-class faculty.

LEARN BY DOING

Practice processes and methods through simulations, assessments, case studies, and tools.

LEARN FROM OTHERS

Connect with an international community of professionals interested in solving complex problems.

LEARN ON DEMAND

Access all of the content online and watch videos on the go.

REFLECT AND APPLY

Bring your new skills to your organization, through examples from technical work environments and ample prompts for reflection.

DEMONSTRATE YOUR SUCCESS

Earn a Professional Certificate and 5 Continuing Education Units (CEUs) from MIT.

LEARN FROM THE BEST

Access cutting edge, research-based multimedia content developed by MIT professors & industry experts.

THIS PROGRAM IS FOR YOU IF...

You have a bachelor's degree in engineering (e.g., mechanical, civil, aerospace, chemical, materials, nuclear, biological, electrical, etc.) or the physical sciences.

You have proficient knowledge of college-level mathematics including differential calculus, linear algebra, and statistics.

You have some experience with MATLAB (R). Programming experience is not necessary, but knowledge of MATLAB (R) is very useful.

Your industry is or will be impacted by machine learning.

Prepare for this program with free online   resources.

JUSTIFY YOUR PROFESSIONAL DEVELOPMENT

Many companies offer professional development benefits to their employees but sometimes starting the conversation is the hardest part of the process.

Use these talking points, stats, and email template to advocate for your professional development through MIT xPRO's online professional certificate program Machine Learning, Modeling, and Simulation: Engineering Problem-Solving in the Age of AI.  

What Learners Are Saying

Mit xpro learners are not only scientists, engineers, technicians, managers and consultants – they are change agents. they take the initiative, push boundaries, and define the future..

Vivian D'Souza, Model Based Systems Analysis Engineer at Dana Incorporated

"The course was a fantastic blend of concepts and practical applications. Professor Youssef's content is unmatched to other similar courses that I've tried, and not to mention his enthusiasm for the topic is contagious."

Rachael Naoum, Product Definition Engineer at Dassault Systems

"I loved this course. At first, I was a bit intimidated, it's been a while since I've done any hardcore math. However, the layout of this course made it super easy to follow along with all the concepts and I felt very well-guided throughout the graded assignments. I like how the lectures are broken into short videos for each topic, making it easy to replay and digest. Very well-thought-out course."

Tolga Kaya, Professor of Electrical and Computer Engineering at Sacred Heart University

"This course allowed me to dig deeper [into] the foundations of machine learning and the underlying mechanism of the main algorithms that are used. As a MATLAB user, I particularly appreciated the utilization of MATLAB instead of straight black box python libraries."

Bill Bear, Agile Transformation Coach

"Great course for learning the concepts and methods behind machine learning! The course was prepared and delivered in a thoughtful way that provided good challenges and plenty of helpful information. This is just the kind of result that I have come to expect with MIT xPRO."

Juharasha Shaik, Senior Staff Software Engineer at Visa

"This course has helped me gain more understanding of the various algorithms that can be applied to the problems that we face during data analysis and modeling. I definitely recommend this course [for a] great understanding of the available tools/algorithms/methods to analyze various use cases and help model the solution."

MIT FACULTY

Youssef M. Marzouk

Faculty Co-Director of MIT Center of Computational Engineering, Professor of Aeronautics & Astronautics and Director of Aerospace Computational Design Laboratory, MIT

George Barbastathis

Professor of Mechanical Engineering, MIT

Heather Kulik

Associate Professor of Chemical Engineering, MIT

John Williams

Professor Civil & Environmental Engineering, MIT

Themistoklis Sapsis

Associate Professor of Mechanical & Ocean Engineering, MIT

Markus Buehler

McAfee Professor of Engineering & Head, Department of Civil & Environmental Engineering, MIT

Richard Braatz

Edwin R. Gilliland Professor of Chemical Engineering, MIT

Justin Solomon

Associate Professor of Electrical Engineering and Computer Science, MIT

Laurent Demanet

Professor of Applied Mathematics & Director of MIT's Earth Resources Laboratory

NOT READY TO ENROLL?

Watch a free demo video..

Machine Learning offers important new capabilities for solving today’s complex problems, but it’s not a panacea. To get beyond the hype, engineers and scientists must discern how and where machine learning tools are the best option — and where they are not.

Submit your information in the form above and watch a short demo video on the online program — what makes it different from other machine learning courses, what you'll learn, and how you will learn it.

Propel Your Career On Your Terms

Technology is accelerating at an unprecedented pace causing disruption across all levels of business. Tomorrow’s leaders must demonstrate technical expertise as well as leadership acumen in order to maintain a technical edge over the competition while driving innovation in an ever-changing environment.

MIT uniquely understands this challenge and how to solve it with decades of experience developing technical professionals. MIT xPRO’s online learning programs leverage vetted content from world-renowned experts to make learning accessible anytime, anywhere. Designed using cutting-edge research in the neuroscience of learning, MIT xPRO programs are application focused, helping professionals build their skills on the job.

Embrace change. Enhance your skill set. Keep learning. MIT xPRO is with you each step of the way.

Have questions about the program?

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

Download Free PDF

How to Model It—Problem Solving for the Computer Age

1991, Applied Mathematical Modelling

Related papers

Computers & Mathematics with Applications, 1986

Energy and Buildings, 2009

Management and Computing Division at Lincoln University whose research and teaching interests are in computing and quantitative disciplines. Previously this group was the academic section of the Centre for Computing and Biometrics at Lincoln University. The group teaches subjects leading to a Bachelor of Applied Computing degree and a computing major in the Bachelor of Commerce and Management. In addition, it contributes computing, statistics and mathematics subjects to a wide range of other Lincoln University degrees. In particular students can take a computing and mathematics major in the BSc. The ACMS group is strongly involved in postgraduate teaching leading to honours, masters and PhD degrees. Research interests are in modelling and simulation, applied statistics, end user computing, computer assisted learning, aspects of computer networking, geometric modelling and visualisation. Research Reports Every paper appearing in this series has undergone editorial review within the A...

SIAM Journal on Scientific Computing, 2008

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

UNITEXT, 2020

Mathematical and Computer Modelling, 1989

Mathematics and Computers in Simulation, 2006

Notices of the AMS, 2009

Proceedings of the 13th WSEAS …, 2009

Studia Universitatis Babeș-Bolyai Digitalia, 2018

Simulation Practice and Theory, 1997

Computational Methods in Applied Mathematics, 2000

Related topics

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

• Corpus ID: 117972866

How to Model It: Problem Solving for the Computer Age

• A. Starfield , K. Smith , A. Bleloch
• Published 1 April 1994
• Computer Science

209 Citations

Practical solutions for making models indispensable in conservation decision‐making.

• Highly Influenced

How to Build a Course in Mathematical–Biological Modeling: Content and Processes for Knowledge and Skill

Response of breonadia salicina in the sabie river to an extreme flood: implications for demographic modelling, monitoring and adaptive river management in the kruger national park, ten years of individual-based modelling in ecology: what have we learned and what could we learn in the future, k-means clustering as a speciation mechanism within an individual-based evolving predator-prey ecosystem simulation, analysing the mathematical discourse of biology assignments: the case of a graduate fisheries management course, story-thinking, computational-thinking, programming and software engineering, guidance on good practice in conducting scientific assessments in animal health using modelling, agent-based modelling of urban district energy system decarbonisation – a systematic literature review, advancing relevance, credibility, legitimacy, and effectiveness as a heuristic for local-parallel scenarios, one reference, related papers.

Showing 1 through 3 of 0 Related Papers

Fourier–Gegenbauer pseudospectral method for solving periodic fractional optimal control problems

New citation alert added.

This alert has been successfully added and will be sent to:

You will be notified whenever a record that you have chosen has been cited.

To manage your alert preferences, click on the button below.

New Citation Alert!

Please log in to your account

Information & Contributors

Bibliometrics & citations, view options, index terms.

Computing methodologies

Artificial intelligence

Control methods

Computational control theory

Mathematics of computing

Mathematical analysis

Functional analysis

Approximation

Mathematical optimization

Numerical analysis

Theory of computation

Design and analysis of algorithms

Recommendations

Fourier-gegenbauer pseudospectral method for solving time-dependent one-dimensional fractional partial differential equations with variable coefficients and periodic solutions.

In this paper, we present a novel pseudospectral (PS) method for solving a new class of initial-value problems (IVPs) of time-dependent one-dimensional fractional partial differential equations (FPDEs) with variable coefficients and periodic ...

High-order, stable, and efficient pseudospectral method using barycentric Gegenbauer quadratures

The work reported in this article presents a high-order, stable, and efficient Gegenbauer pseudospectral method to solve numerically a wide variety of mathematical models. The proposed numerical scheme exploits the stability and the well-conditioning of ...

A Simple Accurate Method for Solving Fractional Variational and Optimal Control Problems

We develop a simple and accurate method to solve fractional variational and fractional optimal control problems with dependence on Caputo and Riemann---Liouville operators. Using known formulas for computing fractional derivatives of polynomials, we ...

Information

Published in.

Elsevier Science Publishers B. V.

Netherlands

Publication History

Author tags.

• Fourier collocation
• Fractional optimal control
• Gegenbauer quadrature
• Optimal periodic control
• Periodic fractional derivative
• Pseudospectral method
• Research-article

Contributors

Other metrics, bibliometrics, article metrics.

• 0 Total Citations
• 0 Total Downloads
• Downloads (Last 12 months) 0
• Downloads (Last 6 weeks) 0

View options

Login options.

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

Share this publication link.

Copying failed.

Share on social media

Affiliations, export citations.

• Please download or close your previous search result export first before starting a new bulk export. Preview is not available. By clicking download, a status dialog will open to start the export process. The process may take a few minutes but once it finishes a file will be downloadable from your browser. You may continue to browse the DL while the export process is in progress. Download
• Download citation
• Copy citation

We are preparing your search results for download ...

We will inform you here when the file is ready.

Your file of search results citations is now ready.

Your search export query has expired. Please try again.