About Experience Projects Publications Skills Contact
Incoming MSc · University of Zurich · Sep 2026

Charan Selva Dhanush Ravi

AI Researcher · CV · NLP · GNN · RL · Interpretable Models

Advancing novel neural architectures, graph-based learning, and vision systems across research and applied domains. Heading to UZH Zurich for my Masters in September 2026.

charan — research.json
SCROLL
Research Vision

Building AI That Sees, Reads,
Reasons & Learns.

My research spans Computer Vision, NLP, Graph Neural Networks, Reinforcement Learning, and Interpretable Models — exploring how machines can perceive, understand, and act across complex real-world domains.

Latest Research · Novel Architecture
White Box Polynomial Neural Networks

A novel neural architecture designed from the ground up for complete interpretability. Traditional activation functions are removed entirely — instead, the model is structured so that each layer naturally and necessarily increases the polynomial degree of the entire network. The result: every single computation in the model is analytically extractable. No black box. No approximations. Every bit of detail, on demand.

No activations · Degree grows layer by layer · Fully extractable
PNN Layer Structure
Degree Grows With Depth

Unlike standard networks where depth adds opacity, each layer here raises the expressiveness and the tractability simultaneously.

// Standard network — opaque
h = σ( Wx + b )  // why?
 
// PNN — each layer, +1 degree
Layer 1poly degree k
Layer 2poly degree k+1
Layer npoly degree k+n−1
 
// Every weight, extractable.
🔬
Key Result
Full Extractability

Because the architecture is polynomial by construction, every detail of what the model learned can be extracted analytically — no probing, no surrogate models, no guessing.

🧭
Current Research Areas
Five Active Fronts
Computer Vision
Natural Language Processing
Graph Neural Networks
Reinforcement Learning
Interpretable Models
Current Focus
Human Behaviour Modelling
Active Research

Applying the PNN framework to model and extract interpretable representations of complex human behavioural patterns.

01

About Me

🎓
Incoming · Masters Degree
University of Zurich — UZH
MSc  ·  Zurich, Switzerland  ·  September 2026
Confirmed · Sep 2026
// Exploring AI across vision, language, graphs, policy & beyond.

I'm a final-year B.Tech student in Computer Science (AI & ML) at Chennai Institute of Technology, with a CGPA of 8.2/10. I'll be joining the University of Zurich (UZH) for my Masters in September 2026. My research spans Computer Vision, NLP, Graph Neural Networks, Reinforcement Learning, and Interpretable Models — exploring how machines can perceive, understand, reason, and act across complex real-world problems.

I've had the privilege of conducting research at NIT Tiruchirappalli (Multimodal VQA, IoT, Virtual Surgery) and Universiti Tunku Abdul Rahman, Malaysia (Stock Forecasting, Transformers).

My current original research on White Box Polynomial Neural Networks is a novel architecture built without any traditional activation functions — the model's structure ensures each layer increases the polynomial degree of the entire network, making every learned relationship analytically extractable and fully interpretable.

Outside research: Smart India Hackathon 2025 Finalist and active language learner (German A1 — Goethe Institut).

0
Research Internships
0
Publications
0
CGPA / 10
0
Major Projects
02

Research Experience

Research Intern — 2 Terms
05/2024 – 05/2025
National Institute Of Technology Tiruchirappalli
Tiruchirappalli, India
  • Term I: Multimodal VQA — transformer pipeline providing auditory assistance to visually impaired individuals.
  • Term II: Soil saturation automation — IoT system using ESP32 for real-time remote monitoring and control of moisture levels.
  • Virtual Surgery interface — OCR-based tumour extraction and placement in Unity 3D anatomy simulations.
Research Intern
07/2024 – 09/2024
Universiti Tunku Abdul Rahman (UTAR)
Perak, Malaysia
  • Comprehensive review of ML/DL models for stock market forecasting — identifying research gaps and published at ICSGRC 2025.
  • Designed a multimodal Transformer model with attention mechanisms for enhanced stock trend prediction accuracy.
Undergraduate Researcher
10/2022 – Present
Centre for Additive Manufacturing
Chennai, India
  • Analysed 3D printing parameters and their effect on material properties; curated datasets for ML pipelines.
  • LSTM-based deep neural networks achieving >95% accuracy in compressive strength prediction.
  • Integrated AI into manufacturing; contributed to patents and peer-reviewed publications.
03

Featured Projects

🌱
02/2025 – 05/2025
Remote Sensing & Soil Moisture Monitoring
IoT system using ESP32 to sense, control and remotely monitor soil moisture — maintaining optimum saturation for structural foundation integrity.
IoTESP32PythonCivil AI
💧
10/2024 – 12/2024
Water Supply & Demand Forecasting
Dual LSTM models with >95% accuracy for regional water forecasting. Interactive dashboard for feature-output trend analysis.
LSTMPyTorchTime SeriesDashboard
👁️
05/2024 – 08/2024
Visual Bot — Multimodal VQA
Transformer-based VQA model giving auditory assistance to visually impaired individuals. Fine-tuned with ULMFiT transfer learning.
TransformersVQAULMFiTNLP
🖨️
02/2023 – 05/2023
Layer Height Monitoring (3D Printing)
Computer vision pipeline detecting additive manufacturing anomalies in real time. Hybrid PyTorch model achieving >90% accuracy.
Computer VisionPyTorchOpenCVAdditive Mfg.
04

Publications

2025
ICSGRC 2025
Deep analysis on ML/DL techniques for stock forecasting. Proposed a Transformer + agent-based learning framework to improve prediction accuracy and profit margins.
2023
Materials Today Communication
Examined 3D printer parameters on compressive strength and developed LSTM deep architecture achieving over 95% prediction accuracy.
05

Skills & Technologies

Specialisations
Interpretable AIPNNComputer VisionNLPGraph Neural NetworksReinforcement LearningLLMsMultimodal AITransformers
ML Frameworks
PyTorchTensorFlowHuggingFaceOpenCVMatplotlib
Languages
PythonCJava
Data Analysis
NumPyPandasDBMSTableau
Web Dev
ReactJSFlask
Hardware / IoT
ESP32Unity 3DIoT Systems
06

Awards & Certifications

Awards
🏆
Smart India Hackathon 2025
Finalist · Government of India · 20 Dec 2024
Certifications
📊
Data Analysis
Google
🧠
Deep Learning Nanodegree
Udacity
🇩🇪
German A1
Goethe Institut
07

Get In Touch

Open to Research & Collaborations

Whether you want to discuss AI research, novel architectures, or potential collaborations — or just want to connect before I head to UZH — my inbox is always open.

Currently Available For

Actively seeking opportunities in research and applied AI.

Research Internships
ML Collaborations
Full-time (2026)
Open Source
Send a Message