Machine Learning Blog | ML@CMU | Carnegie Mellon University

computer vision deep learning machine learning Research

Tracking Any Pixel in a Video

by Adam Harley / September 9, 2022

We upgrade pixels into PIPs: “Persistent Independent Particles”. With this representation, we track any pixel over time, and overcome visibility issues with a learned temporal prior. Motion estimation is a fundamental task of computer vision, with extremely broad applications. By tracking something, you can build models of its various properties: shape, texture, articulation, dynamics, affordances, and so on. More fine-grained tracking allows more fine-grained understanding. For robots, fine-grained tracking also enables fine-grained manipulation. Even setting aside downstream AI-related applications, motion…

43 7408

machine learning Research

Long-term Dynamics of Fairness Intervention in Connection Recommender Systems

September 2, 2022

Figure 1: Modeled recommendation cycle. We demonstrate how enforcing group-fairness in every recommendation slate separately does not necessarily promote equity in second order variables of interest like network size. Connection recommendation is at the heart of user experience in many…

36 1411

machine learning reinforcement learning Research

Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPs

August 26, 2022

Figure 1. Our implementation of recurrent model-free RL outperforms the on-policy version (PPO/A2C-GRU), and a recent model-based POMDP algorithm (VRM) on most tasks of a POMDP benchmark where VRM was evaluated in their paper. While algorithms for decision-making typically focus…

43 1986

machine learning Research

Galaxies on Graph Neural Networks

August 19, 2022

The current accelerated expansion of the Universe is driven by the mysterious dark energy. Upcoming astronomical imaging surveys, such as LSST at Rubin Observatory, are set to provide unprecedented precise measurement of cosmological parameters, including this dark energy, using measurements…

37 3547

Educational machine learning

auton-survival: An Open-Source Package for Regression, Counterfactual Estimation, Evaluation and Phenotyping Censored Time-to-Event Data

August 5, 2022

Real-world decision-making often requires reasoning about when an event will occur. The overarching goal of such reasoning is to help aid decision-making for optimal triage and subsequent intervention. Such problems involving estimation of Times-to-an-Event frequently arise across multiple…

65 3464

artificial intelligence automl deep learning machine learning Research

Does AutoML work for diverse tasks?

July 7, 2022

Over the past decade, machine learning (ML) has grown rapidly in both popularity and complexity. Driven by advances in deep neural networks, ML is now being applied far beyond its traditional domains like computer vision and text processing, with applications…

77 4911 1

computer vision deep learning machine learning Research

Deep Attentive Variational Inference

May 27, 2022

Figure 1: Overview of a local variational layer (left) and an attentive variational layer (right) proposed in this post. Attention blocks in the variational layer are responsible for capturing long-range statistical dependencies in the latent space of the hierarchy. Generative…

90 6172

machine learning Research

An Experimental Design Perspective on Model-Based Reinforcement Learning

May 6, 2022

Reinforcement learning (RL) has achieved astonishing successes in domains where the environment is easy to simulate. For example, in games like Go or those in the Atari library, agents can play millions of games in the course of days to…

104 8550

Research

Two Experiments in Peer Review: Posting Preprints and Citation Bias

April 1, 2022

There is increasing interest in computer science and elsewhere to understand and improve peer review (see here for an overview). With this motivation, we conducted two experiments regarding peer review which we summarize in this blog post.…

242 8800 3

machine learning Research

Assessing Generalization of SGD via Disagreement

March 4, 2022

Imagine training a deep network twice with two different random seeds on the same data, and then measuring the rate at which they disagree on unlabeled test points. Naively, they can disagree with one another with probability anywhere between zero…

178 9006

Older Posts

Machine Learning Blog | ML@CMU | Carnegie Mellon University

Statistics:

Categories: