Innovative Projects Exploring Neural Network Capabilities in Deep Learning
Explore cutting-edge project ideas in the realm of deep learning, ranging from training neural networks to understand symbolic relationships and mathematical operations to utilizing neural power units for combined symbolic and numerical data. Discover datasets for mathematical reasoning and American Sign Language studies, along with resources for fake news detection and text summarization.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
Project Ideas IST 597: Foundation of Deep Learning Spring 2023
Can Neural Networks understand Symbolic Relationships Consider a Circular Ordered Sequence of symbols: Define Operations : * such that: * gives the next symbol in sequence e.g. Train NN to learn these relationships under given set of symbols Train neural networks to generate sequence for a given Grammar
Can Neural Networks learn operations on numbers (+ / - / x / ) Here the input to the neural network is the two numbers and the operation. The output should be the result E.g. Input : 12 , 13, + ; Output : 25 What about fractional input? Can we extend this to solve numerical expressions? Can we learn NN in compositional way Graph NN vs TreeNN vs Transformers ? 2ndOrder RNNs and Memory Modules?
Can Neural Networks understand Comparisons? (> / < / ==, bonus : >=, <=) 23 > 4, 5 == 5 etc, What about math expressions? Compare number of objects in an image: >
Neural Power Units for combined Symbolic and numeral Data https://arxiv.org/pdf/2006.01681.pdf Can we reduce a mathematical expression using Neural Power Units? Are these units compositional ?
Datasets: Mathematical Reasoning: https://artofproblemsolving.com/community/c3158_usa_contests https://dlmf.nist.gov https://github.com/deepmind/mathematics_dataset https://math-qa.github.io https://www.cs.rit.edu/~dprl/ARQMath/arqmath-resources.html
Datasets American Sign Language Dataset: https://www.kaggle.com/datasets/grassknoted/asl-alphabet https://www.kaggle.com/datasets/datamunge/sign-language-mnist YELP Dataset https://www.kaggle.com/datasets/yelp-dataset/yelp-dataset Fake News Detection https://paperswithcode.com/datasets?task=fake-news-detection
Datasets: Summerization: https://github.com/allenai/scitldr https://paperswithcode.com/datasets?q=summarization&v=lst&o=match https://huggingface.co/datasets?task_categories=task_categories:summariza tion&sort=downloads Hugging Face: https://huggingface.co/datasets ACL Anthologies https://github.com/shauryr/ACL-anthology-corpus
Project Ideas: Stanford CS230 past projects: https://cs230.stanford.edu/past-projects/ CMU 11-785 past projects: https://deeplearning.cs.cmu.edu/shared/project.html
