Zobrazeno 1 - 10
of 105
pro vyhledávání: '"SULLIVAN, ALAN P."'
Autor:
Ota, Kei, Tung, Hsiao-Yu, Smith, Kevin A., Cherian, Anoop, Marks, Tim K., Sullivan, Alan, Kanezaki, Asako, Tenenbaum, Joshua B.
The world is filled with articulated objects that are difficult to determine how to use from vision alone, e.g., a door might open inwards or outwards. Humans handle these objects with strategic trial-and-error: first pushing a door then pulling if t
Externí odkaz:
http://arxiv.org/abs/2210.12521
In this paper, we present InSeGAN, an unsupervised 3D generative adversarial network (GAN) for segmenting (nearly) identical instances of rigid objects in depth images. Using an analysis-by-synthesis approach, we design a novel GAN architecture to sy
Externí odkaz:
http://arxiv.org/abs/2108.13865
Autor:
Ota, Kei, Jha, Devesh K., Romeres, Diego, van Baar, Jeroen, Smith, Kevin A., Semitsu, Takayuki, Oiki, Tomoaki, Sullivan, Alan, Nikovski, Daniel, Tenenbaum, Joshua B.
Humans quickly solve tasks in novel systems with complex dynamics, without requiring much interaction. While deep reinforcement learning algorithms have achieved tremendous success in many complex tasks, these algorithms need a large number of sample
Externí odkaz:
http://arxiv.org/abs/2011.07193
In this paper we propose integrating a priori knowledge into both design and training of convolutional neural networks (CNNs) to learn object representations that are invariant to affine transformations (i.e., translation, scale, rotation). According
Externí odkaz:
http://arxiv.org/abs/1909.00114
Face super-resolution methods usually aim at producing visually appealing results rather than preserving distinctive features for further face identification. In this work, we propose a deep learning method for face verification on very low-resolutio
Externí odkaz:
http://arxiv.org/abs/1903.10974
We propose a novel {\it Equilibrated Recurrent Neural Network} (ERNN) to combat the issues of inaccuracy and instability in conventional RNNs. Drawing upon the concept of autapse in neuroscience, we propose augmenting an RNN with a time-delayed self-
Externí odkaz:
http://arxiv.org/abs/1903.00755
In this paper we study the problem of convergence and generalization error bound of stochastic momentum for deep learning from the perspective of regularization. To do so, we first interpret momentum as solving an $\ell_2$-regularized minimization pr
Externí odkaz:
http://arxiv.org/abs/1903.00760
Autor:
van Baar, Jeroen, Sullivan, Alan, Cordorel, Radu, Jha, Devesh, Romeres, Diego, Nikovski, Daniel
Learning robot tasks or controllers using deep reinforcement learning has been proven effective in simulations. Learning in simulation has several advantages. For example, one can fully control the simulated environment, including halting motions whi
Externí odkaz:
http://arxiv.org/abs/1809.04720
Autor:
Cherian, Anoop, Sullivan, Alan
Unpaired image-to-image translation is the problem of mapping an image in the source domain to one in the target domain, without requiring corresponding image pairs. To ensure the translated images are realistically plausible, recent works, such as C
Externí odkaz:
http://arxiv.org/abs/1807.04409