Sitemap
A list of all the posts and pages found on the site. For you robots out there is an XML version available for digesting as well.
Pages
Posts
Future Blog Post
Published:
This post will show up by default. To disable scheduling of future posts, edit config.yml and set future: false.
Blog Post number 4
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 3
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 2
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
Blog Post number 1
Published:
This is a sample blog post. Lorem ipsum I can’t remember the rest of lorem ipsum and don’t have an internet connection right now. Testing testing testing this blog post. Blog posts are cool.
papers
Structured Latent Variable Models for Articulated Object Interaction
Published:
This paper explores how a robot can learn a low-dimensional representation of doors from videos of them opening or closing, enabling the inference of door-related parameters and interaction outcomes. Instead of relying solely on labeled datasets, the study employs a semi-supervised approach using the Neural Statistician, a structured latent variable model that separates shared context-level variables (common across all images of the same door) from instance-level variables (specific to each image). The model effectively generates realistic door image embeddings, which outperform context-free baselines in tasks such as predicting door parameters and optimizing actions in a visual bandit door-opening scenario, demonstrating its utility for more efficient and accurate robotic interaction.
Recommended citation: Structured Latent Variable Models for Articulated Object Interaction. Emily Liu, Michael Noseworthy, Nicholas Roy. 2023.
Download Paper
Semi-Supervised Neural Processes for Articulated Object Interaction
Published:
This paper addresses the challenge of limited labeled action data for robotic object manipulation by introducing the Semi-Supervised Neural Process (SSNP). SSNP combines small amounts of labeled interaction data with abundant unlabeled visual data, using a jointly trained reward-prediction and autoencoding framework to extract task-relevant features. This approach reduces the need for extensive retraining and computational resources while improving generalization. The model outperforms other semi-supervised methods in a door-opening task, achieving superior performance with significantly less data.
Recommended citation: Semi-Supervised Neural Processes for Articulated Object Interaction. Emily Liu, Michael Noseworthy, Nicholas Roy. RSS 2024 Workshop on Structural Priors as Inductive Biases for Learning Robot Dynamics.
Download Paper | Download Slides
Leveraging Intermediate Neural Collapse with Simplex ETFs for Efficient Deep Neural Networks
Published:
This paper examines the phenomenon of neural collapse, where network activations, class means, and linear classifier weights converge to a simplex equiangular tight frame (ETF) during training, enhancing interpretability, robustness, and generalization. Building on findings that neural collapse extends beyond the final layer in fully connected networks, the authors propose two novel methods: Adaptive-ETF, which enforces simplex ETF constraints across all layers beyond a certain depth, and ETF-Transformer, which applies these constraints to feedforward layers in transformer blocks. Both methods maintain performance while significantly reducing trainable parameters, offering efficient alternatives for network training and regularization.
Recommended citation: Leveraging Intermediate Neural Collapse with Simplex ETFs for Efficient Deep Neural Networks (2024). Emily Liu. NeurIPS 2024 Workshop on Mathematics of Modern Machine Learning.
Download Paper
portfolio
Degree Distribution Bias in Graph Neural Networks
6.867: Machine Learning, Fall 2021
Interaction-Based Material Recognition with Multimodal Audio-Visual Datasets
6.869: Machine Vision, Spring 2022
Impact of Simplex Equiangular Tight Frames on Intermediate Neural Collapse and Effective Depth in Deep Neural Networks
9.520: Statistical Learning Theory, Fall 2023
Noisy-channel Bayesian Inference in Mandarin Listening Tasks
9.66: Computational Cognitive Science, Fall 2022
Somos Sequences and the Cube Recurrence
18.204: Seminar in Discrete Mathematics, Spring 2023
The Slepian-Wolf Theorem and Error Probabilities in Correlated Sources
18.424: Seminar in Information Theory, Fall 2023
Controlled FastComposer
6.5940: TinyML and Efficient Deep Learning, Fall 2023
Benchmarking Model and Hardware Architectures
6.5931: Hardware for Deep Learning, Spring 2024
teaching
Introduction to Cellular Automata
Workshop, Splash at MIT, 2022
Cellular automata are dynamic systems of “cell”-like objects that evolve over time to generate patterns. They have been studied extensively for their application to physics, biology, ecology, sociology, and more, in addition to posing interesting theoretical questions. Ideal for lovers of programming, mathematics, and art, this class will open with a brief lecture on history and application, as well as the rules for some basic CA. Students will then have the opportunity to implement elementary CA and Conway’s Game of Life in Python.
Randomized Algorithms
High school summer course, MIT High School Studies Program, 2023
Randomized algorithms are algorithms that have access to a random source during computation, meaning that the output of the algorithm is non-deterministic. Oftentimes, randomized algorithms are useful in reducing computation time needed for otherwise slow or complex tasks. How do we design randomized algorithms to speed up computation without trading off correctness of our output? In this course, we will cover several Las Vegas and Monte Carlo algorithms, such as randomized quicksort, Frievald’s algorithm, game tree evaluation, and primality testing algorithms.
Sentiment Analysis Primer
Workshop, HackMIT, 2023
This workshop serves as an introduction to sentiment and text analysis. It begins with an overview of neural network fundamentals and sentiment analysis, explaining its core concepts, applications, and significance in understanding text. Participants are then introduced to the Hugging Face ecosystem, highlighting its role in providing easy access to pre-trained NLP models.