CDS Master’s students have a unique opportunity to solve real-world problems through the Capstone course in the final year of their program. The capstone course is designed to apply knowledge into practice and to develop and improve critical skills such as problem-solving and collaboration skills.
Students are encouraged to collaborate with research labs within the NYU community and with industry partners to investigate pressing issues, applying data science to the following areas:
- Probability and statistical analyses.
- Natural language processing.
- Big Data analysis and modeling.
- Machine learning and computational statistics.
- Coding and software engineering.
- Visualization modeling.
- Neural networks.
- Signal processing.
- High dimensional statistics.
Capstone projects present students with the opportunity to work in their field of interest and gain exposure to applicable solutions. Project sponsors, NYU labs, and external partners, in turn receive the benefit of having a new perspective applied to their projects.
“Capstone is a unique opportunity for students to solve real world problems through projects carried out in collaboration with industry partners or research labs within the NYU community,” says Capstone advisor and CDS Research Fellow Anastasios Noulas. “It is a vital experience for students ahead of their graduation and prior to entering the market, as it helps them improve their skills, especially in problem solving contexts that are atypical compared to standard courses offered in the curriculum. Cooperation within teams is another crucial skill built through the Capstone experience as projects are typically run across groups of 3 or 4 people”.
The Capstone Project offers the opportunity for organizations to propose a project that our graduate students will work on as part of their curriculum for one semester. Information on the course along with a questionnaire to propose a project, can be found on the Capstone Fall 2020 Project Submission Form. If you have any questions please reach out to Loraine Nascimento at LN38@nyu.edu.
Best Fall 2019 Capstone Posters
Infering the Topic(s) of Wikipedia Articles
By Marina Zavalina, Sarthak Agarwal, Chinmay Singhal, Peeyush Jain
Option Portfolio Replication and Hedging in Deep Reinforcement Learning
By Bofei Zhang, Jiayi Du, Yixuan Wang, Muyang Jin
Featured NYU Capstone Projects
Deep Learning for Breast Cancer Detection
By Jason Phang, Jungkyu (JP) Park, Thibault Fevry, Zhe Huang, The B-Team
Featured Industry Capstone Projects
Predicting Stock Market Movements using Public Sentiment Data & Sequential Deep Learning Models
Accern Corporation Summary
2019 Capstone Project List
- Adversarial Attacks Against Linear and Deep-learning Regressions in Astronomy
- Automated Breast Cancer Screening
- Automatic Legal Case Summaries
- Beer NLP
- Cross-task Transfer Between Language Understanding Tasks in NLP
- Dark Matter and Stellar Stream Detection using Deep Learned Clustering
- Exploiting Google Street View to Generate Global-scale Data Sets for Training Next Generation Cyber-Physical Systems
- Federated Incremental Learning
- Fraud Detection in Monetary Transactions Between Bank Accounts
- Guided Image Upsampling
- Improving State of the Art Cross-Lingual Word-Embeddings
- Inferring the Topic(s) of Wikipedia Articles
- Latent Semantic Topics Distribution Over Web Content Corpus
- Lease Renewal Probability Prediction
- Machine Learning for Adaptive Fuzzy String Matching
- Market Segmentation from Retailer Behavior
- Modeling the Experienced Dental Curriculum from Student Data
- Modelling NBA Games
- Movie Preference Prediction
- MRI Image Reconstruction
- NLP Metalearning
- Predict next sales office location
- Predicting Stock Market Movements using Public Sentiment Data & Sequential Deep Learning Models
- Predictive Maintenance Techniques
- Reinforcement Learning for Replication and Hedging of Option
- Self-supervised Machine Listening
- Sentence Classification of TripAdvisor ‘Points-of-Interest’ Reviews
- Simulating the Dark Matter Distribution of the Universe with Deep Learning
- SMaPP2: Joint Embedding of User-content and Network Structure to Enable a Common coordinate that captures ideology, geography and user topic spectrum.”
- Sparse Deconvolution Methods for Microscopy Imaging Data Analysis
- Stereotype and Unconscious Bias in Large Datasets
- Structuring Exploring and Exploiting NIH’s Clinical Trials Database
- The Analysis, Visualization, and Understanding of Big Urban Noise Data
- Unsupervised and Self-supervised Learning for Medical Notes
- Unsupervised Generative Video Dubbing
- Using Deep Generative Models to de-noise Noisy Astronomical Data
Other Past Capstone Projects
- “Active Physical Inference via Reinforcement Learning”
- “Deep Multi-Modal Content-User Embeddings for Music Recommendation”
- “Fluorescent Microscopy Image Restoration”
- “Learning Visual Embeddings for Reinforcement Learning”
- “Offensive Speech Detection on Twitter”
- “Predicting Movement Primitives in Stroke Patients using IMU Sensors”
- “Recurrent Policy Gradients For Smooth Continuous Control”
- “The Quality-Quantity Tradeoff in Deep Learning”
- “Trend Modeling in Childhood Obesity Prediction”
- “Twitter Food/Activity Monitor”