A national research priority
Scientific machine learning supports national initiatives in artificial intelligence, advanced computing and interdisciplinary research, with applications in climate, energy, healthcare and engineering.
Rajiv Gandhi Institute of Petroleum Technology (RGIPT)
An Institute of National Importance
Under the Ministry of Petroleum and Natural Gas, Government of India
Jais, Amethi, Uttar Pradesh, India
An ANRF-sponsored five-day workshop on
Recent Advances in Computational Mathematics and Scientific Machine Learning
Theory, Methods and Applications
RACM–SciML 2026
Dates12–16 Dec 2026
ModeHybrid
VenueDepartment of Mathematical Sciences
RGIPT Jais, Amethi, Uttar Pradesh, India
RGIPT Jais, Amethi, Uttar Pradesh, India
–days
–hours
–minutes
to the inaugural session
Why now
Mathematics in the age of artificial intelligence
In the past two years, artificial intelligence has progressed from solving standard exercises to contributing to research in mathematics. This development is changing how mathematical problems are explored, proved and verified.
AI can now generate conjectures, constructions and proofs at great speed. However, establishing that a result is correct, stable and meaningful continues to depend on rigorous mathematical analysis.
Relevance of this workshop
A similar change is taking place in applied mathematics, where neural networks are now used to solve differential equations and to model physical systems. RACM–SciML 2026 addresses this area directly.
1
Mathematical rigour
Neural network solvers can produce results quickly, but their reliability must be established. Convergence, stability and error analysis provide the basis for this.
2
Classical and machine learning methods
Finite element, spectral and meshfree methods are studied alongside physics-informed neural networks and neural operators, with attention to the strengths and limitations of each.
3
Hands-on
Participants implement and test physics-informed solvers in Python under the guidance of the resource persons.
4
Future research directions
Scientific machine learning offers open problems in theory, algorithms and applications to energy, fluid dynamics and biology, providing research opportunities for students and young faculty.
About the workshop
Mathematics and scientific machine learning workshop at RGIPT
RACM–SciML 2026 is an ANRF-sponsored workshop at RGIPT Jais, Amethi, Uttar Pradesh, India, connecting computational mathematics with scientific machine learning. Held in hybrid mode from 12–16 December 2026, it covers physics-informed neural networks (PINNs), numerical methods for differential equations and operator learning, with hands-on Python sessions for students, researchers and faculty.
Neural networks are increasingly used to solve differential equations. Convergence theory, stability and error analysis remain essential for judging the reliability of their solutions.
This five-day hybrid workshop, organised by the Department of Mathematical Sciences, RGIPT, brings the two worlds together. Participants revisit the analytical and numerical foundations of differential equations, then see how neural-network-based solvers extend, complement and sometimes challenge them.
Mornings are lectures on theory and convergence from leading researchers. Afternoons are hands-on coding sessions in Python, so you leave able to implement, test and critically assess both classical schemes and physics-informed solvers on problems of your own.
Lectures by leading researchers
Lectures by experts from IISc, IITs, Brown University and WUR on finite elements, domain decomposition, PINNs and operator learning.
Hands-on sessions
Guided laboratory sessions in NumPy, PyTorch/TensorFlow and DeepXDE. Participants develop working notebooks that can be extended for theses and research projects.
New research directions
Discussions with speakers and peers help students and young faculty identify open research questions.
The hosts
About the organisers
Rajiv Gandhi Institute of Petroleum Technology (RGIPT)
RGIPT was established by an Act of Parliament in 2007 as an Institute of National Importance. It is an energy-focused institute under the Ministry of Petroleum and Natural Gas, Government of India, offering B.Tech., M.Tech., MBA and Ph.D. programmes across engineering, management and the sciences.
2007Act of Parliament
3campuses: Jais, Bengaluru, Sivasagar
INIInstitution of National Importance
The Jais campus in Amethi hosts the workshop. Its research facilities include a Central Instrumentation Facility, a Design and Innovation Center, and a Centre of Excellence for AI/ML in the petroleum and energy sector, a natural home for scientific machine learning.
Visit rgipt.ac.in →
Department of Mathematical Sciences
The department became a full-fledged department in 2021 and offers undergraduate, postgraduate and doctoral programmes, including the B.Tech. in Mathematics and Computing (since 2021–22), which trains students in computation, mathematical and statistical modelling, and simulation. It also teaches mathematics and statistics to every other department of the institute.
Visit the department website →
Organisers
Organising committee
Patron
Prof. Harish HiraniDirector, RGIPT
Chairperson
Dr. Manoj Kumar RajpootHead, Department of Mathematical Sciences
Convener
Dr. Soniya DhamaAssistant Professor, Department of Mathematical Sciences
Organising committee members
Prof. Chanchal KunduDepartment of Mathematical Sciences
Dr. Dhrubasish BhattacharyyaDepartment of Mathematical Sciences
Dr. Gobinda RakshitDepartment of Mathematical Sciences
Dr. Pradeep DasDepartment of Mathematical Sciences
Dr. Sudeep KunduDepartment of Mathematical Sciences
Dr. Saurabh BansalDepartment of Mathematical Sciences
Advisory and technical committee
Prof. P. V. S. N. MurthyIIT Kharagpur
Prof. Jitendra KumarIIT Ropar
Dr. Amit Kumar VermaIIT Patna
Dr. Kapil KantIIIT Gwalior
Themes
What we will cover
The programme covers these topics, but is not limited to them.
- Nonlinear ODEs, PDEs and boundary value problems
- Quasilinearization and Green's function based methods
- Finite difference, finite element, spectral and meshfree RBF methods
- Singular, singularly perturbed, delay and stochastic problems
- Convergence, stability and error analysis
- Neural networks and automatic differentiation
- Physics-informed neural networks: theory, training and pitfalls
- Adaptive sampling, loss balancing and domain decomposition
- RBF-PINNs and sparse PINNs for stiff and delay problems
- Operator learning: DeepONet and Fourier neural operators
- Hybrid methods, reduced-order models and equation discovery
- Uncertainty quantification; applications in energy, fluids and biology
Learning pathways: numerical analysis to AI for science
The numerical analysis workshop sessions explore scientific computing, numerical methods for differential equations, boundary value problems and finite element methods. Discussions of convergence, stability and error analysis connect mathematical theory with reliable computation.
The physics-informed machine learning sessions introduce neural network solvers for differential equations, automatic differentiation, PINN training and domain decomposition. The operator learning sessions cover DeepONet, Fourier neural operators and hybrid numerical–neural methods.
Hands-on Python sessions support practical learning with PyTorch and DeepXDE. Students, research scholars and faculty interested in an AI for science workshop can connect these methods to applications in energy, fluid dynamics and engineering. Online and on-campus participation make this a hybrid scientific computing workshop in Uttar Pradesh.
Hands-on withPythonNumPyPyTorchTensorFlowDeepXDE
Resource persons
Our Distinguished Speakers (Tentative)
Explore the profiles, research interests and selected contributions of our distinguished speakers.
GK
Prof. George Em Karniadakis
Charles Pitts Robinson and John Palmer Barstow Professor of Applied Mathematics
Brown University, USA
Scientific machine learning; physics-informed neural networks; stochastic multiscale modelling
View profile & short bio
Prof. Karniadakis develops computational methods that connect applied mathematics, scientific computing and machine learning. He received his Ph.D. from MIT and was elected to the US National Academy of Engineering in 2022 for contributions spanning computational algorithms and applications to complex physical systems.
Ranked #1 worldwide in mathematics in Research.com’s 2026 ranking, as reported by Brown University. See Google Scholar for current citation figures.
Areas of interest
- Physics-informed neural networks and scientific machine learning
- Numerical methods for differential equations
- Stochastic modelling and uncertainty quantification
RP
Prof. Rajesh K. Pandey
Professor, Department of Mathematical Sciences
IIT (BHU), Varanasi, India
Numerical PDEs; wavelets; fractional differential equations; image processing and AI
View profile & short bio
Prof. Rajesh Kumar Pandey works at the intersection of numerical analysis, fractional differential equations and image processing. He obtained his doctorate from the Institute of Technology, Banaras Hindu University, and is a Professor of Mathematical Sciences at IIT (BHU), Varanasi.
Areas of interest
- Numerical methods for partial differential equations
- Wavelets
- Fractional differential equations
- Image processing and artificial intelligence
TG
Prof. Thirupathi Gudi
IISc Bangalore
Scientific computing, finite element methods
View profileSN
Prof. Srinivasan Natesan
Professor (HAG), Department of Mathematics
IIT Guwahati, India
Singular perturbations; PINNs; fractional equations; domain decomposition; CFD
View profile & short bio
Prof. Srinivasan Natesan is Professor (HAG) in the Department of Mathematics at IIT Guwahati. His work brings numerical analysis and parallel computation together with applications in differential equations, scientific machine learning and multiscale modelling.
Areas of interest
- Numerical methods for singular perturbation problems: finite difference, finite element and finite volume methods
- Big data analytics, artificial intelligence, machine learning, deep learning and PINNs
- Computational finance
- Efficient numerical methods for fractional-order differential equations
- Homogenization and multiscale problems; Bloch wave decomposition
- Domain decomposition and parallel computation of differential equations
- Computational fluid dynamics
BK
Prof. B. V. Rathish Kumar
Professor, Department of Mathematics
IIT Kanpur, India
Numerical methods for PDEs; numerical analysis; scientific computing; fluid dynamics; AI/ML
View profile & short bio
Prof. B. V. Rathish Kumar is a Professor in the Department of Mathematics at IIT Kanpur. His research connects numerical analysis and scientific computing with the modelling and simulation of fluid flows, including artificial intelligence and machine-learning methods for partial differential equations.
Areas of interest
- Numerical methods for partial differential equations
- Numerical analysis
- Scientific computing
- Fluid dynamics
- Artificial intelligence and machine learning
SK
Prof. Sanjeev Kumar
Professor, Department of Mathematics; Joint Faculty, Mehta Family School of Data Science and AI
IIT Roorkee, India
Mathematical image processing; computational imaging; machine learning
View profile & short bio
Prof. Sanjeev Kumar develops mathematical and computational approaches to imaging and machine learning at IIT Roorkee. His interests span image recovery, secure imaging and three-dimensional reconstruction, alongside learning-based methods for image analysis.
Areas of interest
- Mathematical image processing and computational algorithms for image restoration
- Image encryption and secret sharing
- Quantum imaging and 3-D reconstruction
- Neural networks and tree-based classifiers
- Machine learning applications in image processing
PP
Dr. Puneet Pasricha
IIT Ropar
Mathematical finance, ML in finance, applied probability, stochastic processes
View profileAV
Dr. Amit K. Verma
Associate Professor, Department of Mathematics
IIT Patna, India
Numerical analysis; differential equations; scientific machine learning
View profile & short bio
Dr. Amit K. Verma is an Associate Professor of Mathematics at IIT Patna. His research includes the analysis and numerical solution of nonlinear differential equations and boundary value problems, with interests extending to scientific machine learning.
Areas of interest
- Numerical analysis
- Differential equations and nonlinear boundary value problems
- Scientific machine learning
TK
Dr. Taniya Kapoor
Assistant Professor, Artificial Intelligence Group
Wageningen University & Research, Netherlands
Machine learning for science; physics-informed modelling; scientific machine learning
View profile & short bio
Dr. Taniya Kapoor develops machine-learning models informed by differential equations and physical principles to address scientific and engineering challenges. She earned her Ph.D. at TU Delft and subsequently held an AI Schmidt Postdoctoral Fellowship at the University of Oxford.
Areas of interest
- Machine learning for the sciences and science for machine learning
- Integrating PDEs, ODEs and physical laws into machine-learning models
- Physics-informed modelling for global scientific and engineering challenges
Speaking at the workshop? Session format, travel, stay and the speaker information form are in the speakers' corner.
Speakers' corner
Programme
Daily schedule
Tentative programme. Select a day to see the full session timings. Talk titles and speakers will be added as they are confirmed.
Day 1Saturday12 Dec 2026Inauguration · foundations of nonlinear differential equationsView details →
Day 2Sunday13 Dec 2026Numerical methods · first hands-on sessionView details →
Day 3Monday14 Dec 2026Physics-informed neural networks · hands-on sessionView details →
Day 4Tuesday15 Dec 2026Operator learning and hybrid methods · hands-on sessionView details →
Day 5Wednesday16 Dec 2026Applications and future directions · valedictoryView details →
Before you arrive
Get ready for the hands-on labs
Labs use Python. Prepare your laptop before the workshop so you can follow the hands-on sessions.
1Easiest: Google Colab
Nothing to install. Sign in to Google Colab with any Google account and check that a notebook opens. Bring a laptop and charger.
2Or install locally
Install Python 3.10+ (Anaconda or python.org), then run:
pip install numpy scipy matplotlib torch deepxde jupyter
3Refresh the basics
A first course in ODEs and PDEs, and basic Python (arrays, loops, plotting), is all you need. Recordings, notes, slides and notebooks will be shared where provided or permitted by the speakers.
Registration
Important dates
- 1 Oct 2026Registration opens
- 20 Nov 2026Last date to register
- Within one week after registrationSelection intimation
- 12–16 Dec 2026Workshop
Fees and how to register
Registration is open.
Seats are limited. Selection is intimated within one week after registration. The offline fee includes a registration kit, food during the workshop, hostel accommodation and a participation certificate. Guest-house accommodation is subject to availability and institute norms.
UG / PG student (Indian nationals)₹1,000offline
Research scholar (Indian nationals)₹1,500offline
Postdoc / faculty (Indian nationals)₹2,000offline
Industry / R&D (Indian nationals)₹3,000offline
Indian nationals₹500online
Foreign nationalsUS$30online or offline
To complete your registration, kindly pay the applicable fee and fill in the registration form with your transaction ID, payment receipt, and a valid institutional or government-issued ID card. You will receive a registration number and an acknowledgement by email upon submission. Selection is intimated within one week after registration. You will be notified by email. We look forward to welcoming you to the workshop!
What your registration includes
Offline participants: lectures, hands-on sessions, questions and discussions, a registration kit, food during the workshop, hostel accommodation and a participation certificate. Guest-house accommodation is subject to availability and institute norms.
Online participants (Indian nationals: ₹500; foreign nationals: US$30): live access to lectures and hands-on sessions, opportunities to ask questions, and a participation certificate.
For both modes: Recordings, notes, slides and notebooks will be shared where provided or permitted by the speakers.
Refund policy: Registration fees are non-refundable, subject to applicable rules and regulations.
Limited seats
Travel support for offline students
Partial travel expenses will be reimbursed for UG/PG students attending in person. Support is limited and allotted on a first come, first served basis, so register early.
- Register in an offline category and pay the fee.
- Select "Request travel support" in the registration form. Your position in the queue is shown in the acknowledgement email.
- Support is confirmed with your selection intimation.
- Keep your original tickets; reimbursement is processed as per institute norms.
Support is partial and depends on the number of eligible participants.
Let us remind you before registration closes
Planning to register later? Leave your email and we will help you remember the 20 November 2026 deadline.
Check your registration status
Verify a certificate
Enter the certificate ID printed at the bottom of a RACM–SciML 2026 certificate.
Attend in person
Benefits of attending in person
Online participation is available. However, attending in person offers greater opportunities for interaction and learning.
Guidance in hands-on sessions
On-campus participants receive direct assistance from the resource persons during the laboratory sessions.
Interaction with speakers
Breaks and informal sessions allow participants to discuss research, doctoral and postdoctoral opportunities with senior researchers.
Stay on campus
Hostel accommodation is included in the offline registration fee. Guest-house accommodation is subject to availability and institute norms.
Places of historical interest
Ayodhya, Lucknow, Prayagraj, Varanasi and Kanpur are within a few hours of the campus.
To our speakers: we would be pleased to welcome you on campus. Your presence offers participants from across the country a valuable opportunity for direct academic interaction.
Venue
Getting to RGIPT Jais, Amethi, Uttar Pradesh, India
Department of Mathematical Sciences, RGIPT Jais, Amethi, Uttar Pradesh, India
Rajiv Gandhi Institute of Petroleum Technology, Jais, Amethi, Uttar Pradesh 229304, India, an Institute of National Importance established by an Act of Parliament in 2007.
By airLucknow CCS International Airport, about 110 km (2.5–3 h by road). Ayodhya and Prayagraj airports are alternatives with fewer flights.
By railGuru Gorakhnath Dham Railway Station (station code: GUGD), formerly Jais Railway Station, is only about 5 minutes from RGIPT, and major trains stop there. Raebareli Junction and Lucknow Charbagh offer wider connectivity.
By roadThe institute is on the Raebareli–Sultanpur highway. Direct buses run from Delhi to Raebareli, and from Raebareli bus station buses towards Sultanpur pass the institute. Taxis are available from Lucknow and Raebareli.
December weather: pleasant days and cold nights, often below 10 °C, with dense morning fog that can delay trains and flights. Pack warm clothing and keep buffer time.
Distances are approximate road distances; check a map app for your route.
Beyond the workshop
Explore nearby
Jais sits in the heart of Awadh. Five historic cities are within a day's reach; extend your trip by a day or two.
Ayodhya
Temple town on the Saryu
~125 km · 3 h- Shri Ram Janmabhoomi Mandir, the temple complex at the birthplace of Lord Ram.
- Hanuman Garhi, a hilltop temple reached by 76 steps.
- Kanak Bhawan, known for its ornate interiors.
- Ram ki Paidi and Saryu ghats, best at the evening aarti.
Tip: mobile phones and bags are not allowed inside the Ram Mandir; use the lockers and start early to avoid queues.
Lucknow
City of Nawabs
~120 km · 3 h- Bara Imambara and Bhool Bhulaiya, the 18th-century hall and its labyrinth.
- Rumi Darwaza and Chota Imambara, a short walk away.
- Hazratganj for an evening stroll and chikankari shopping.
- Old Lucknow food trail in Chowk and Aminabad: kebabs, kulfi, and makhan malai, a winter-only treat.
Tip: if you fly in, spend your arrival or departure day here.
Prayagraj
Where three rivers meet
~135 km · 3 h- Triveni Sangam, the confluence of the Ganga, Yamuna and mythical Saraswati, by boat.
- Anand Bhawan, the Nehru family home, now a museum.
- Khusro Bagh and the Allahabad Museum.
- University of Allahabad, one of India's oldest universities.
Tip: morning boat rides are calmest; carry a shawl, the river breeze is cold in December.
Varanasi
The oldest living city
~240 km · 5 h- Ganga Aarti at Dashashwamedh Ghat every evening.
- Dawn boat ride along the ghats from Assi to Manikarnika.
- Shri Kashi Vishwanath Dham and its corridor.
- Sarnath (about 10 km away), site of the Buddha's first sermon; and the BHU campus.
Tip: plan it as an overnight trip, ideally after the valedictory session.
Kanpur
Industrial and academic hub
~165 km · 3.5 h- IIT Kanpur campus, a must for students considering research careers.
- JK Temple, a blend of ancient and modern architecture.
- Allen Forest Zoo, one of India's largest open-forest zoos.
- Moti Jheel and the Ganga Barrage for evening walks.
Tip: Kanpur Central is a major rail hub, handy for onward travel west.
Travel times are approximate by road and vary with traffic and fog.
Get in touch
Contact
Email: racmsml2026@gmail.com
Convener: soniyad@rgipt.ac.in
Department of Mathematical Sciences, RGIPT, Jais, Amethi, Uttar Pradesh 229304, India
FAQ
Questions
Who can apply?
UG and PG students, research scholars, postdoctoral fellows, faculty and professionals in mathematics, computational science, engineering and allied disciplines.
Do I need to know machine learning already?
No. A first course in differential equations and basic Python is enough. The workshop builds from foundations to current research.
What should I bring?
A laptop with Python installed for the hands-on sessions, your ID card, and warm clothing.
How do online participants join?
A joining link is emailed to confirmed online participants before the workshop. The online fee (₹500 for Indian nationals; US$30 for foreign nationals) includes live lectures, hands-on sessions, opportunities to ask questions and a participation certificate. Recordings, notes, slides and notebooks will be shared where provided or permitted by the speakers.
Who gets travel support?
Offline UG/PG students, partially and on a first come, first served basis, within the available budget. See travel support.
Is accommodation available?
Hostel accommodation is included in the offline fee. Guest-house accommodation is subject to availability and institute norms. Request accommodation while registering.