Zobrazeno 1 - 10
of 81
pro vyhledávání: '"Sudderth, Erik B."'
Autor:
Ricci, Federica Zoe, Sudderth, Erik B., Lee, Jaylen, Peters, Megan A. K., Vannucci, Marina, Guindani, Michele
We consider the problem of analyzing multivariate time series collected on multiple subjects, with the goal of identifying groups of subjects exhibiting similar trends in their recorded measurements over time as well as time-varying groups of associa
Externí odkaz:
http://arxiv.org/abs/2406.17131
By composing graphical models with deep learning architectures, we learn generative models with the strengths of both frameworks. The structured variational autoencoder (SVAE) inherits structure and interpretability from graphical models, and flexibl
Externí odkaz:
http://arxiv.org/abs/2306.08230
We develop a new framework for learning variational autoencoders and other deep generative models that balances generative and discriminative goals. Our framework optimizes model parameters to maximize a variational lower bound on the likelihood of o
Externí odkaz:
http://arxiv.org/abs/2012.06718
Autor:
Ren, Zhile, Sudderth, Erik B.
We develop new representations and algorithms for three-dimensional (3D) object detection and spatial layout prediction in cluttered indoor scenes. We first propose a clouds of oriented gradient (COG) descriptor that links the 2D appearance and 3D po
Externí odkaz:
http://arxiv.org/abs/1906.04725
We tackle the problem of automatically reconstructing a complete 3D model of a scene from a single RGB image. This challenging task requires inferring the shape of both visible and occluded surfaces. Our approach utilizes viewer-centered, multi-layer
Externí odkaz:
http://arxiv.org/abs/1902.06729
To date, top-performing optical flow estimation methods only take pairs of consecutive frames into account. While elegant and appealing, the idea of using more than two frames has not yet produced state-of-the-art results. We present a simple, yet ef
Externí odkaz:
http://arxiv.org/abs/1810.10066
Autor:
Hughes, Michael C., Hope, Gabriel, Weiner, Leah, McCoy, Thomas H., Perlis, Roy H., Sudderth, Erik B., Doshi-Velez, Finale
Supervisory signals can help topic models discover low-dimensional data representations that are more interpretable for clinical tasks. We propose a framework for training supervised latent Dirichlet allocation that balances two goals: faithful gener
Externí odkaz:
http://arxiv.org/abs/1712.00499
Word2vec (Mikolov et al., 2013) has proven to be successful in natural language processing by capturing the semantic relationships between different words. Built on top of single-word embeddings, paragraph vectors (Le and Mikolov, 2014) find fixed-le
Externí odkaz:
http://arxiv.org/abs/1711.03946
Given two consecutive frames from a pair of stereo cameras, 3D scene flow methods simultaneously estimate the 3D geometry and motion of the observed scene. Many existing approaches use superpixels for regularization, but may predict inconsistent shap
Externí odkaz:
http://arxiv.org/abs/1707.08313
Autor:
Hughes, Michael C., Weiner, Leah, Hope, Gabriel, McCoy Jr., Thomas H., Perlis, Roy H., Sudderth, Erik B., Doshi-Velez, Finale
Supervisory signals have the potential to make low-dimensional data representations, like those learned by mixture and topic models, more interpretable and useful. We propose a framework for training latent variable models that explicitly balances tw
Externí odkaz:
http://arxiv.org/abs/1707.07341