Spaces:
Running
Running
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>NeuralScribe | ML Research Portfolio</title> | |
| <script src="https://cdn.tailwindcss.com"></script> | |
| <script src="https://unpkg.com/feather-icons"></script> | |
| <script src="https://cdn.jsdelivr.net/npm/vanta@latest/dist/vanta.globe.min.js"></script> | |
| <script src="https://cdn.jsdelivr.net/npm/[email protected]/lib/anime.min.js"></script> | |
| <style> | |
| @import url('https://fonts.googleapis.com/css2?family=Space+Grotesk:wght@300;400;500;600;700&display=swap'); | |
| body { | |
| font-family: 'Space Grotesk', sans-serif; | |
| } | |
| .gradient-text { | |
| background: linear-gradient(90deg, #3b82f6 0%, #8b5cf6 50%, #ec4899 100%); | |
| -webkit-background-clip: text; | |
| background-clip: text; | |
| color: transparent; | |
| } | |
| .publication-card:hover { | |
| transform: translateY(-5px); | |
| box-shadow: 0 20px 25px -5px rgba(0, 0, 0, 0.1), 0 10px 10px -5px rgba(0, 0, 0, 0.04); | |
| } | |
| #vanta-bg { | |
| position: absolute; | |
| top: 0; | |
| left: 0; | |
| width: 100%; | |
| height: 100%; | |
| z-index: -1; | |
| } | |
| .timeline-item:before { | |
| content: ''; | |
| position: absolute; | |
| left: -38px; | |
| top: 0; | |
| width: 15px; | |
| height: 15px; | |
| border-radius: 50%; | |
| background: #3b82f6; | |
| } | |
| </style> | |
| </head> | |
| <body class="bg-gray-50 text-gray-800"> | |
| <!-- Vanta.js Background --> | |
| <div id="vanta-bg"></div> | |
| <!-- Navigation --> | |
| <nav class="fixed w-full bg-white/80 backdrop-blur-md z-50 shadow-sm"> | |
| <div class="max-w-6xl mx-auto px-4 sm:px-6 lg:px-8"> | |
| <div class="flex justify-between h-16"> | |
| <div class="flex items-center"> | |
| <span class="text-xl font-bold gradient-text">NeuralScribe</span> | |
| </div> | |
| <div class="hidden md:flex items-center space-x-8"> | |
| <a href="#about" class="text-gray-700 hover:text-blue-600 transition">About</a> | |
| <a href="#research" class="text-gray-700 hover:text-blue-600 transition">Research</a> | |
| <a href="#publications" class="text-gray-700 hover:text-blue-600 transition">Publications</a> | |
| <a href="#experience" class="text-gray-700 hover:text-blue-600 transition">Experience</a> | |
| <a href="#contact" class="px-4 py-2 bg-blue-600 text-white rounded-md hover:bg-blue-700 transition">Contact</a> | |
| </div> | |
| <div class="md:hidden flex items-center"> | |
| <button id="menu-btn" class="text-gray-700"> | |
| <i data-feather="menu"></i> | |
| </button> | |
| </div> | |
| </div> | |
| </div> | |
| </nav> | |
| <!-- Mobile Menu --> | |
| <div id="mobile-menu" class="hidden fixed inset-0 bg-white z-40 pt-20 px-4"> | |
| <div class="flex flex-col space-y-6 text-lg"> | |
| <a href="#about" class="border-b border-gray-100 pb-4">About</a> | |
| <a href="#research" class="border-b border-gray-100 pb-4">Research</a> | |
| <a href="#publications" class="border-b border-gray-100 pb-4">Publications</a> | |
| <a href="#experience" class="border-b border-gray-100 pb-4">Experience</a> | |
| <a href="#contact" class="px-4 py-2 bg-blue-600 text-white rounded-md w-max">Contact</a> | |
| </div> | |
| </div> | |
| <!-- Hero Section --> | |
| <section class="min-h-screen flex items-center pt-16"> | |
| <div class="max-w-6xl mx-auto px-4 sm:px-6 lg:px-8 py-20"> | |
| <div class="grid md:grid-cols-2 gap-12 items-center"> | |
| <div> | |
| <h1 class="text-4xl md:text-5xl lg:text-6xl font-bold mb-6"> | |
| <span class="gradient-text">Optimizing ML Efficiency</span><br> | |
| Through Algorithmic Innovation | |
| </h1> | |
| <p class="text-xl text-gray-600 mb-8"> | |
| PhD candidate specializing in making machine learning models more efficient without requiring larger architectures or more data. | |
| </p> | |
| <div class="flex space-x-4"> | |
| <a href="#research" class="px-6 py-3 bg-blue-600 text-white rounded-md hover:bg-blue-700 transition">Explore Research</a> | |
| <a href="#publications" class="px-6 py-3 border border-gray-300 rounded-md hover:bg-gray-100 transition">View Publications</a> | |
| </div> | |
| </div> | |
| <div class="relative"> | |
| <div class="bg-gradient-to-br from-blue-500 to-purple-600 rounded-2xl p-1"> | |
| <img src="http://static.photos/technology/640x360/42" alt="ML Research" class="rounded-2xl w-full h-auto"> | |
| </div> | |
| <div class="absolute -bottom-6 -right-6 bg-white p-4 rounded-xl shadow-lg"> | |
| <div class="flex items-center"> | |
| <div class="bg-green-100 p-2 rounded-full mr-3"> | |
| <i data-feather="award" class="text-green-600"></i> | |
| </div> | |
| <div> | |
| <p class="font-medium">Hugging Face</p> | |
| <p class="text-sm text-gray-500">Research Award 2023</p> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Research Focus Section --> | |
| <section id="about" class="py-20 bg-white"> | |
| <div class="max-w-6xl mx-auto px-4 sm:px-6 lg:px-8"> | |
| <div class="text-center mb-16"> | |
| <h2 class="text-3xl md:text-4xl font-bold mb-4">Research Philosophy</h2> | |
| <div class="w-20 h-1 bg-gradient-to-r from-blue-500 to-purple-600 mx-auto"></div> | |
| </div> | |
| <div class="grid md:grid-cols-3 gap-8"> | |
| <div class="bg-gray-50 p-6 rounded-xl border border-gray-200"> | |
| <div class="bg-blue-100 w-12 h-12 rounded-full flex items-center justify-center mb-4"> | |
| <i data-feather="cpu" class="text-blue-600"></i> | |
| </div> | |
| <h3 class="text-xl font-semibold mb-3">Efficiency First</h3> | |
| <p class="text-gray-600">Developing techniques that improve model performance without increasing computational requirements or data needs.</p> | |
| </div> | |
| <div class="bg-gray-50 p-6 rounded-xl border border-gray-200"> | |
| <div class="bg-purple-100 w-12 h-12 rounded-full flex items-center justify-center mb-4"> | |
| <i data-feather="layers" class="text-purple-600"></i> | |
| </div> | |
| <h3 class="text-xl font-semibold mb-3">Algorithmic Innovation</h3> | |
| <p class="text-gray-600">Creating novel approaches like GRPO that fundamentally change how models learn from data.</p> | |
| </div> | |
| <div class="bg-gray-50 p-6 rounded-xl border border-gray-200"> | |
| <div class="bg-pink-100 w-12 h-12 rounded-full flex items-center justify-center mb-4"> | |
| <i data-feather="globe" class="text-pink-600"></i> | |
| </div> | |
| <h3 class="text-xl font-semibold mb-3">Real-World Impact</h3> | |
| <p class="text-gray-600">Applying research to domains like molecular science and computer vision with measurable improvements.</p> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Detailed Research Section --> | |
| <section id="research" class="py-20 bg-gray-50"> | |
| <div class="max-w-6xl mx-auto px-4 sm:px-6 lg:px-8"> | |
| <div class="text-center mb-16"> | |
| <h2 class="text-3xl md:text-4xl font-bold mb-4">Research Highlights</h2> | |
| <div class="w-20 h-1 bg-gradient-to-r from-blue-500 to-purple-600 mx-auto"></div> | |
| </div> | |
| <div class="grid md:grid-cols-2 gap-12 mb-20"> | |
| <div> | |
| <h3 class="text-2xl font-bold mb-4 gradient-text">Group Relative Policy Optimization (GRPO)</h3> | |
| <p class="text-gray-600 mb-6"> | |
| A novel reinforcement learning framework that achieves superior performance with reduced computational overhead compared to traditional PPO methods. GRPO introduces group-based relative rewards that stabilize training while maintaining sample efficiency. | |
| </p> | |
| <div class="space-y-4"> | |
| <div class="flex items-start"> | |
| <div class="bg-blue-100 p-2 rounded-full mr-4"> | |
| <i data-feather="check-circle" class="text-blue-600"></i> | |
| </div> | |
| <div> | |
| <h4 class="font-medium">20% Faster Convergence</h4> | |
| <p class="text-sm text-gray-500">Compared to standard PPO in Atari benchmarks</p> | |
| </div> | |
| </div> | |
| <div class="flex items-start"> | |
| <div class="bg-purple-100 p-2 rounded-full mr-4"> | |
| <i data-feather="check-circle" class="text-purple-600"></i> | |
| </div> | |
| <div> | |
| <h4 class="font-medium">Reduced Variance</h4> | |
| <p class="text-sm text-gray-500">More stable training dynamics across environments</p> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| <div class="bg-white p-6 rounded-xl shadow-md border border-gray-100"> | |
| <img src="http://static.photos/science/640x360/23" alt="GRPO Visualization" class="rounded-lg mb-4 w-full"> | |
| <div class="flex justify-between items-center"> | |
| <span class="text-sm font-medium text-gray-500">Research Visualization</span> | |
| <a href="#" class="text-blue-600 hover:underline flex items-center"> | |
| View Paper <i data-feather="arrow-right" class="ml-1 w-4 h-4"></i> | |
| </a> | |
| </div> | |
| </div> | |
| </div> | |
| <div class="grid md:grid-cols-2 gap-12"> | |
| <div class="order-last md:order-first"> | |
| <img src="http://static.photos/education/640x360/57" alt="Molecular Science" class="rounded-xl shadow-md w-full"> | |
| </div> | |
| <div> | |
| <h3 class="text-2xl font-bold mb-4 gradient-text">Molecular Science Applications</h3> | |
| <p class="text-gray-600 mb-6"> | |
| Developed specialized architectures for molecular property prediction that outperform traditional methods while using significantly fewer parameters. Our approach combines geometric deep learning with efficient attention mechanisms. | |
| </p> | |
| <div class="space-y-4"> | |
| <div class="flex items-start"> | |
| <div class="bg-green-100 p-2 rounded-full mr-4"> | |
| <i data-feather="bar-chart-2" class="text-green-600"></i> | |
| </div> | |
| <div> | |
| <h4 class="font-medium">15% Accuracy Improvement</h4> | |
| <p class="text-sm text-gray-500">On QM9 benchmark dataset</p> | |
| </div> | |
| </div> | |
| <div class="flex items-start"> | |
| <div class="bg-yellow-100 p-2 rounded-full mr-4"> | |
| <i data-feather="zap" class="text-yellow-600"></i> | |
| </div> | |
| <div> | |
| <h4 class="font-medium">3× Faster Inference</h4> | |
| <p class="text-sm text-gray-500">Compared to baseline models</p> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </div> | |
| </section> | |
| <!-- Publications Section --> | |
| <section id="publications" class="py-20 bg-white"> | |
| <div class="max-w-6xl mx-auto px-4 sm:px-6 lg:px-8"> | |
| <div class="text-center mb-16"> | |
| <h2 class="text-3xl md:text-4xl font-bold mb-4">Publications</h2> | |
| <div class="w-20 h-1 bg-gradient-to-r from-blue-500 to-purple-600 mx-auto"></div> | |
| </div> | |
| <div class="grid md:grid-cols-2 gap-6"> | |
| <!-- Publication 1 --> | |
| <div class="publication-card bg-gray-50 p-6 rounded-xl border border-gray-200 transition duration-300"> | |
| <div class="flex justify-between items-start mb-3"> | |
| <span class="bg-blue-100 text-blue-800 text-xs font-medium px-2.5 py-0.5 rounded">NeurIPS 2023</span> | |
| <div class="flex space-x-2"> | |
| <a href="#" class="text-gray-400 hover:text-blue-600"> | |
| <i data-feather="file-text" class="w-4 h-4"></i> | |
| </a> | |
| <a href="#" class="text-gray-400 hover:text-blue-600"> | |
| <i data-feather="github" class="w-4 h-4"></i> | |
| </a> | |
| </div> | |
| </div> | |
| <h3 class="text-xl font-semibold mb-2">Group Relative Policy Optimization for Sample-Efficient RL</h3> | |
| <p class="text-gray-600 mb-4">Introducing GRPO, a novel RL algorithm that achieves superior performance with reduced computational overhead through group-based relative rewards.</p> | |
| <div class="flex flex-wrap gap-2"> | |
| <span class="bg-gray-100 text-gray-800 text-xs px-3 py-1 rounded-full">Reinforcement Learning</span> | |
| <span class="bg-gray-100 text-gray-800 text-xs px-3 py-1 rounded-full">Efficiency</span> | |
| <span class="bg-gray-100 text-gray-800 text-xs px-3 py-1 rounded-full">PPO</span> | |
| </div> | |
| </div> | |
| <!-- Publication 2 --> | |
| <div class="publication-card bg-gray-50 p-6 rounded-xl border border-gray-200 transition duration-300"> | |
| <div class="flex justify-between items-start mb-3"> | |
| <span class="bg-purple-100 text-purple-800 text-xs font-medium px-2.5 py-0.5 rounded">ICML 2023</span> | |
| <div class="flex space-x-2"> | |
| <a href="#" class="text-gray-400 hover:text-blue-600"> | |
| <i data-feather="file-text" class="w-4 h-4"></i> | |
| </a> | |
| <a href="#" class="text-gray-400 hover:text-blue-600"> | |
| <i data-feather="github" class="w-4 h-4"></i> | |
| </a> | |
| </div> | |
| </div> | |
| <h3 class="text-xl font-semibold mb-2">Efficient Geometric Deep Learning for Molecular Property Prediction</h3> | |
| <p class="text-gray-600 mb-4">A parameter-efficient architecture combining geometric deep learning with attention mechanisms for molecular science applications.</p> | |
| <div class="flex flex-wrap gap-2"> | |
| <span class="bg-gray-100 text-gray-800 text-xs px-3 py-1 rounded-full">Molecular Science</span> | |
| <span class="bg-gray-100 text-gray-800 text-xs px-3 py-1 rounded-full">GNNs</span> | |
| <span class="bg-gray-100 text-gray-800 text-xs px-3 py-1 rounded-full">Efficiency</span> | |
| </div> | |
| </div> | |
| <!-- Publication 3 --> | |
| <div class="publication-card bg-gray-50 p-6 rounded-xl border border-gray-200 transition duration-300"> | |
| <div class="flex justify-between items-start mb-3"> | |
| <span class="bg-green-100 text-green-800 text-xs font-medium px-2.5 py-0.5 rounded">AAAI 2023</span> | |
| <div class="flex space-x-2"> | |
| <a href="#" class="text-gray-400 hover:text-blue-600"> | |
| <i data-feather="file-text" class="w-4 h-4"></i> | |
| </a> | |
| <a href="#" class="text-gray-400 hover:text-blue-600"> | |
| <i data-feather="github" class="w-4 h-4"></i> | |
| </a> | |
| </div> | |
| </div> | |
| <h3 class="text-xl font-semibold mb-2">Data-Efficient Training of Vision Transformers</h3> | |
| <p class="text-gray-600 mb-4">Novel techniques to train vision transformers with limited data while maintaining competitive performance on benchmark datasets.</p> | |
| <div class="flex flex-wrap gap-2"> | |
| <span class="bg-gray-100 | |
| </body> | |
| </html> |