Research & Education

ttt-rl is a reinforcement learning example implemented in C, designed to teach the basics of reinforcement learning through a tic-tac-toe game. The neural network learns to play against a random adversary from scratch, without any pre-existing knowledge of the game. It uses a simple architecture with a single hidden layer and is contained in under 400 lines of C code, with no external libraries. This project is particularly valuable for programmers, especially young programmers, who want to understand new fields through small, self-contained, and well-commented C programs. It demonstrates how RL can learn complex behaviors from basic reward signals.

Responsible AI is an educational tool hosted on Hugging Face Spaces, designed to help users learn about and develop ethical AI systems. It focuses on understanding AI ethics and identifying bias within artificial intelligence applications. The tool is intended for students, educators, and AI enthusiasts who are keen on building AI solutions with a strong foundation in ethical considerations. While the current live website indicates a runtime error, the tool's purpose is clearly to provide resources and a platform for exploring the critical aspects of responsible AI development.

Learn Creole with Jorge is a comprehensive language learning platform designed to make mastering Haitian Creole accessible and enjoyable for all ages. The tool offers a variety of engaging features, including interactive games that transform learning into play, and a rich collection of music and videos that immerse users in Haitian culture. Learners can track their progress with badges and streaks, staying motivated as they build their skills. Whether the goal is to connect with family, deepen a relationship with a Haitian partner, prepare for travel to Haiti, or simply enjoy the process of learning a new language, Learn Creole with Jorge provides a fun and meaningful pathway to fluency.

Coder is an AI software developer tool designed to streamline and enhance the process of AI software development and code generation. This tool is particularly useful for individuals looking to learn coding or automate various software-related tasks. It operates by loading a given website within a border-less, full-screen iframe, providing a seamless and immersive development environment. The tool is hosted on Hugging Face Spaces, indicating its accessibility and potential for community-driven enhancements. Coder aims to simplify complex development workflows, making it an asset for both novice and experienced developers in the AI domain.

ns3-gym is an open-source framework designed to bridge the gap between reinforcement learning (RL) and network simulation. It integrates the popular OpenAI Gym toolkit with the ns-3 network simulator, which is widely used in academic and industry studies for networking protocols and communication technologies. This integration allows researchers to apply RL techniques to complex networking problems, such as cognitive radio channel selection and TCP congestion control. The framework provides a flexible C++ interface within ns-3 to define observation spaces, action spaces, rewards, and game-over conditions, making it highly customizable for various research scenarios. It supports both C++ and Python for agent development and offers examples for quick setup and experimentation.

ml-videos is an actively maintained collection of video resources specifically curated for machine learning enthusiasts, researchers, and students. This repository compiles recorded talks from various machine learning conferences, workshops, seminars, and summer schools, offering a centralized hub for educational content. Building upon an original compilation from a blog post, it aims to provide a comprehensive overview of significant discussions and presentations in the field. Users are encouraged to contribute to the list by submitting pull requests, ensuring its continued growth and relevance. The collection highlights key archives like VideoLectures.NET, TechTalks.tv, and Slideslive, along with specific YouTube channels, making it an invaluable resource for staying updated on machine learning advancements.

GhostPen is an AI text humanizer designed to transform AI-generated content into natural, human-like writing. It features a deep-optimized anti-detection engine that helps users bypass AI detection tools such as Turnitin, GPTZero, and Originality.ai. The tool works by restructuring syntax and perplexity to mimic human writing patterns, making AI text undetectable without altering its original meaning. GhostPen supports various content types including academic papers, blogs, emails, and reports. It emphasizes user privacy, stating that no data logs are kept and text is permanently incinerated after processing, with no AI training on user data. The tool offers a free trial and various paid plans with different word counts and features.

WritingBench is a comprehensive benchmark tool designed for evaluating generative writing models. Users can upload Excel files containing evaluation results, which the application then processes to generate interactive leaderboards, detailed performance tables, and heat-maps. This allows for a clear visualization and comparison of different model performances, highlighting strengths and weaknesses. Hosted on Hugging Face Spaces, WritingBench aims to provide a standardized and accessible platform for researchers and developers to assess and improve their AI writing models. The tool is free to use and offers a structured approach to understanding the nuances of generative writing outputs.

udlbook is an open-source GitHub repository dedicated to the book "Understanding Deep Learning" by Simon J.D. Prince. It serves as a comprehensive educational resource, providing a wealth of materials for students and educators alike. The repository includes Jupyter notebooks that allow for interactive learning and experimentation with deep learning models, as well as course slides for lectures and presentations. Additionally, it contains PDF figures, errata, and an answer booklet for students, making it a complete package for studying deep learning. The content is designed to offer a thorough understanding of deep learning principles, from foundational concepts to advanced topics, and is freely available under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International Public License.

Machine-Learning-Specialization-Coursera is a comprehensive resource offering solutions and notes for the Machine Learning Specialization by Stanford University and Deeplearning.ai, taught by Prof. Andrew Ng on Coursera. This GitHub repository is designed to assist students in reviewing and solidifying their understanding of machine learning principles. It covers all three courses in the specialization, including Supervised Machine Learning (Regression and Classification), Advanced Learning Algorithms, and Unsupervised Learning, Recommenders, and Reinforcement Learning. The repository provides detailed notes, practice quiz solutions, and programming assignment solutions, making it an invaluable companion for anyone undertaking this popular machine learning course.

The machine-learning-specialization repository on GitHub serves as an Open Source resource for individuals engaged in machine learning education and practice. It provides a collection of datasets specifically curated for various machine learning courses, including those from Coursera's Machine Learning Specialization. Users can access datasets such as `amazon_baby`, `home_data`, `image_test_data`, `image_train_data`, `people_wiki.gl`, and `song_data` for Course 1, and `kc_house_data.gl.zip` for Course 2. This repository is designed to support students and learners by offering practical data for implementing and testing machine learning algorithms.

EconML is a Python package developed by Microsoft Research as part of the ALICE (Automated Learning and Intelligence for Causation and Economics) project. It provides a toolkit for estimating heterogeneous treatment effects from observational data, integrating advanced machine learning techniques with econometrics. The package is designed to measure the causal effect of treatment variables on an outcome, controlling for various features, and how this effect varies. It supports methods like Double Machine Learning, Causal Forests, Orthogonal Random Forests, and Meta-Learners, offering flexibility in modeling effect heterogeneity while preserving causal interpretation and providing confidence intervals. EconML is built on standard Python packages for Machine Learning and Data Analysis, making it accessible for data scientists and researchers.

vec2text is an open-source library providing utilities for decoding deep representations, such as sentence embeddings, back into text. It enables users to train various architectures that reconstruct text sequences from embeddings and also run pre-trained models. The library supports both direct inversion from embeddings and inversion of text strings, with options to refine results through multiple steps and increased search space. It is particularly useful for researchers and developers working with text embeddings and language models, offering functionalities like interpolation of embeddings and detailed guidance on training custom inversion and corrector models.

EntitySeg is an open-source toolbox designed for advanced image segmentation tasks, focusing on open-world and high-quality segmentation. It consolidates several cutting-edge algorithms developed by the qqlu group, including Open-World Entity Segmentation (TPAMI2022), High Quality Segmentation for Ultra High-resolution Images (CVPR2022), CA-SSL: Class-Agnostic Semi-Supervised Learning (ECCV2022), and High-Quality Entity Segmentation (ICCV2023 Oral). The toolbox is built using Python and PyTorch, making it accessible for researchers and developers in the computer vision domain. It aims to provide a unified platform for various image segmentation challenges, with future plans to merge all projects for enhanced interoperability and support.

Taboo AI offers an engaging and fun way to learn English and enhance wordplay skills through an interactive game inspired by Taboo. Users can play with public topics contributed by a global community or explore endless possibilities with AI-generated topics. The platform also provides "Daily Lingo," a free personal English learning companion that delivers a new word, definition, and example directly to your phone every day. This tool is recognized as a valuable language learning resource, particularly for English Language Learners (ELLs), as highlighted by educational experts.

Deep-Learning-with-TensorFlow-book is an open-source educational resource designed for individuals looking to learn deep learning using the TensorFlow 2.0 framework. This comprehensive book combines theoretical concepts with practical, real-world case studies, making it highly suitable for beginners. The repository includes a downloadable PDF ebook, complete with a detailed table of contents, along with all the accompanying source code. Additionally, it offers supplementary courseware, including IPython Notebooks for interactive learning. The resource has been widely recognized, adopted by numerous universities as a textbook or reference material, and has received acclaim from authoritative media outlets.

pysc2-examples offers a collection of Deep Reinforcement Learning examples specifically designed for StarCraft II. Built upon Deepmind's pysc2, OpenAI's baselines, and Blizzard's s2client-proto, it provides a robust framework for developers and researchers. The project leverages TensorFlow 1.3 and includes examples for tasks like 'CollectMineralShards' using Deep Q Networks and A2C algorithms. Users can quickly set up the environment, install necessary libraries like pysc2 and baselines, download StarCraft II maps, and then train and enjoy their AI agents. It supports various parameters for training, including algorithm choice (deepq, a2c), total timesteps, exploration fraction, and options for prioritized replay or dueling networks.

Super OCRs Demo is an AI tool hosted on Hugging Face Spaces, designed for experimenting with various small Optical Character Recognition (OCR) models. Users can upload an image and choose from four different OCR engines to process it. Optionally, a custom prompt can be added to guide the recognition process. The application returns the recognized text or markdown. For the DeepSeek model specifically, it also provides a visual output showing the image with highlighted recognized areas, offering a clear understanding of the OCR's performance. This tool is ideal for researchers, developers, and anyone interested in evaluating and comparing different OCR technologies.

Qvantify automates qualitative research by using AI to conduct remote interviews, allowing organizations to gather hundreds of deep feedback interviews daily. The platform enables continuous discovery, market expansion assessment, and solution validation by letting AI interact with customers and users. It combines the simplicity of surveys with the depth of in-depth interviews, offering features like white-labeling for brand customization and localization for conducting interviews in over a hundred languages. Qvantify also provides a REST API for data access and future features include a customer portal, voice input, and outbound calling, making it a comprehensive tool for scalable qualitative feedback.

HUB21 offers premium online tutoring services specifically designed for students pursuing GCSE, A Level, and IB qualifications. The platform connects students with highly qualified tutors from prestigious UK universities such as Oxford, Cambridge, UCL, and Imperial. Lessons are personalized and conducted one-to-one online, focusing on subjects including Maths, Physics, Chemistry, Biology, and Computer Science. Beyond academic support, HUB21 emphasizes mentorship, guiding students to build confidence, master complex subjects, and prepare for admission to leading universities worldwide. The service aims to provide tailored pathways to academic excellence, combining deep subject expertise with proven study strategies for top exam results.

Real3DPortrait is an open-source project providing a PyTorch implementation for one-shot realistic 3D talking portrait synthesis. It allows users to generate high-quality talking face videos from a single source image and a driving audio or video. The tool supports both audio-driven and video-driven methods for generating expressive 3D portraits. Key features include the ability to control mouth amplitude, map initial poses, and provide custom background images. It offers a command-line interface, a Gradio WebUI, and a Google Colab notebook for inference, making it accessible for various users. The project also provides training code for its audio-to-motion and image-to-plane models.

papers-notebook is an open-source GitHub repository dedicated to compiling reading notes on academic papers, primarily in the fields of distributed systems, virtualization, and machine learning. The project aims to document the author's thoughts on papers, including their core ideas, implementation methods, and personal evaluations, with each note ideally under 1000 words. While initially maintained as Markdown files, the project has transitioned to using GitHub issues for ongoing maintenance and welcomes community contributions for paper suggestions. The repository covers a wide range of topics within its focus areas, including schedulers like Mesos and Borg, consensus algorithms like Raft and Paxos, storage systems, virtualization technologies, and security aspects.

linkedin-skill-assessments-quizzes offers a full reference of answers for various LinkedIn skill assessments, including topics like AWS Lambda, REST API, JavaScript, React, Git, HTML, jQuery, MongoDB, Java, Go, Python, machine learning, and PowerPoint. This open-source GitHub repository serves as a valuable resource for individuals looking to prepare for or review answers related to these skill assessments. While LinkedIn skill assessments were discontinued in December 2023, the repository remains a useful learning and reference tool for mastering knowledge in these technical domains.

IqraEval Shared Task @ ArabicNLP 2025 is a specialized AI application hosted on Hugging Face designed for the automatic assessment of Qur’anic recitations. This tool helps users identify and correct mispronunciations by comparing their audio recordings against a database of reference phoneme sequences. It provides detailed feedback, highlighting specific errors to aid in learning and improvement. The application is particularly valuable for researchers, educators, and students involved in Arabic language processing and Qur'anic studies, offering a practical solution for evaluating pronunciation accuracy. Its focus on a specific domain makes it a unique resource for those dedicated to the precise recitation of the Qur'an.