Zobrazeno 1 - 10
of 36
pro vyhledávání: '"Turek, Javier S."'
Autor:
Pink, Mathis, Vo, Vy A., Wu, Qinyuan, Mu, Jianing, Turek, Javier S., Hasson, Uri, Norman, Kenneth A., Michelmann, Sebastian, Huth, Alexander, Toneva, Mariya
Current LLM benchmarks focus on evaluating models' memory of facts and semantic relations, primarily assessing semantic aspects of long-term memory. However, in humans, long-term memory also includes episodic memory, which links memories to their con
Externí odkaz:
http://arxiv.org/abs/2410.08133
Autor:
Chien, Hsiang-Yun Sherry, Turek, Javier S., Beckage, Nicole, Vo, Vy A., Honey, Christopher J., Willke, Ted L.
Sequential information contains short- to long-range dependencies; however, learning long-timescale information has been a challenge for recurrent neural networks. Despite improvements in long short-term memory networks (LSTMs), the forgetting mechan
Externí odkaz:
http://arxiv.org/abs/2105.05944
Publikováno v:
International Conference on Learning Representations 2021
Language models must capture statistical dependencies between words at timescales ranging from very short to very long. Earlier work has demonstrated that dependencies in natural language tend to decay with distance between words according to a power
Externí odkaz:
http://arxiv.org/abs/2009.12727
Autor:
Turek, Javier S., Jain, Shailee, Vo, Vy, Capota, Mihai, Huth, Alexander G., Willke, Theodore L.
Recent work has shown that topological enhancements to recurrent neural networks (RNNs) can increase their expressiveness and representational capacity. Two popular enhancements are stacked RNNs, which increases the capacity for learning non-linear f
Externí odkaz:
http://arxiv.org/abs/1909.00021
Autor:
Mandal, Shantanu, Anderson, Todd A., Turek, Javier S., Gottschlich, Justin, Zhou, Shengtian, Muzahid, Abdullah
Publikováno v:
Proceedings of Machine Learning and Systems (MLSys), 3 (2021), 139-155
The problem of automatic software generation is known as Machine Programming. In this work, we propose a framework based on genetic algorithms to solve this problem. Although genetic algorithms have been used successfully for many problems, one criti
Externí odkaz:
http://arxiv.org/abs/1908.08783
Autor:
Anderson, Michael J., Tamir, Jonathan I., Turek, Javier S., Alley, Marcus T., Willke, Theodore L., Vasanawala, Shreyas S., Lustig, Michael
Magnetic resonance imaging is capable of producing volumetric images without ionizing radiation. Nonetheless, long acquisitions lead to prohibitively long exams. Compressed sensing (CS) can enable faster scanning via sub-sampling with reduced artifac
Externí odkaz:
http://arxiv.org/abs/1809.04195
Autor:
Turek, Javier S., Huth, Alexander
Geodesic distance matrices can reveal shape properties that are largely invariant to non-rigid deformations, and thus are often used to analyze and represent 3-D shapes. However, these matrices grow quadratically with the number of points. Thus for l
Externí odkaz:
http://arxiv.org/abs/1705.10887
Autor:
Zhang, Hejia, Chen, Po-Hsuan, Chen, Janice, Zhu, Xia, Turek, Javier S., Willke, Theodore L., Hasson, Uri, Ramadge, Peter J.
There is a growing interest in joint multi-subject fMRI analysis. The challenge of such analysis comes from inherent anatomical and functional variability across subjects. One approach to resolving this is a shared response factor model. This assumes
Externí odkaz:
http://arxiv.org/abs/1609.09432
Autor:
Chen, Po-Hsuan, Zhu, Xia, Zhang, Hejia, Turek, Javier S., Chen, Janice, Willke, Theodore L., Hasson, Uri, Ramadge, Peter J.
Finding the most effective way to aggregate multi-subject fMRI data is a long-standing and challenging problem. It is of increasing interest in contemporary fMRI studies of human cognition due to the scarcity of data per subject and the variability o
Externí odkaz:
http://arxiv.org/abs/1608.04846
Autor:
Anderson, Michael J., Capotă, Mihai, Turek, Javier S., Zhu, Xia, Willke, Theodore L., Wang, Yida, Chen, Po-Hsuan, Manning, Jeremy R., Ramadge, Peter J., Norman, Kenneth A.
The scale of functional magnetic resonance image data is rapidly increasing as large multi-subject datasets are becoming widely available and high-resolution scanners are adopted. The inherent low-dimensionality of the information in this data has le
Externí odkaz:
http://arxiv.org/abs/1608.04647