QL
Qianli Liao
cs.LGcs.NEstat.MLcs.CVcs.AIModels, Neurologicalq-bio.NCapproximationcomputational neurosciencedeep learning
On Valency
published · living versionsW_fbsjdgcn·v1 · currentpublished
Bridging the Gaps Between Residual Learning, Recurrent Neural Networks and Visual Cortex
with Tomaso Poggio
1 version
Preprints & journals
26 papers in the corpus · 2013–2025Hierarchical Reasoning Models: Perspectives and Misconceptions2510.00355v2 · Renee Ge, Qianli Liao, Tomaso Poggio2025 · 0 citationsarXiv
Self-Assembly of a Biologically Plausible Learning Circuit2412.20018v1 · Qianli Liao, Liu Ziyin, Yulu Gan et al.2024 · 0 citationsarXiv
Dynamics in Deep Classifiers Trained with the Square Loss: Normalization, Low Rank, Neural Collapse, and Generalization Bounds.37223467 · Xu, Mengjia, Rangamani, Akshay, Liao, Qianli et al.2023 · 11 citationsResearch (Washington, D.C.). 2023;6:0024
Hierarchically Compositional Tasks and Deep Convolutional Networks2006.13915v3 · Arturo Deza, Qianli Liao, Andrzej Banburski et al.2020 · 4 citationsarXiv
Theoretical issues in deep networks.32518109 · Poggio, Tomaso, Banburski, Andrzej, Liao, Qianli2021 · 168 citationsProceedings of the National Academy of Sciences of the United States of America. 2020;117(48):30039-30045
Bridging the Gaps Between Residual Learning, Recurrent Neural Networks and Visual Cortex1604.03640v2 · Qianli Liao, Tomaso Poggio2016 · 185 citationsarXivon Valency
Explicit regularization and implicit bias in deep network classifiers trained with the square loss2101.00072v1 · Tomaso Poggio, Qianli Liao2020 · 7 citationsarXiv
Complexity control by gradient descent in deep networks.32094327 · Poggio, Tomaso, Liao, Qianli, Banburski, Andrzej2020 · 35 citationsNature communications. 2020;11(1):1027
Theory III: Dynamics and Generalization in Deep Networks1903.04991v5 · Andrzej Banburski, Qianli Liao, Brando Miranda et al.2019 · 5 citationsarXiv
Theoretical Issues in Deep Networks: Approximation, Optimization and Generalization1908.09375v1 · Tomaso Poggio, Andrzej Banburski, Qianli Liao2019 · 19 citationsarXiv
Biologically-plausible learning algorithms can scale to large datasets1811.03567v3 · Will Xiao, Honglin Chen, Qianli Liao et al.2018 · 47 citationsarXiv
A Surprising Linear Relationship Predicts Test Performance in Deep Networks1807.09659v1 · Qianli Liao, Brando Miranda, Andrzej Banburski et al.2018 · 21 citationsarXiv
View-Tolerant Face Recognition and Hebbian Learning Imply Mirror-Symmetric Neural Tuning to Head Orientation.27916522 · Leibo, Joel Z, Liao, Qianli, Anselmi, Fabio et al.2018 · 57 citationsCurrent biology : CB. 2017;27(1):62-67
Theory IIIb: Generalization in Deep Networks1806.11379v1 · Tomaso Poggio, Qianli Liao, Brando Miranda et al.2018 · 14 citationsarXiv
Theory of Deep Learning III: explaining the non-overfitting puzzle1801.00173v2 · Tomaso Poggio, Kenji Kawaguchi, Qianli Liao et al.2017 · 48 citationsarXiv
Theory of Deep Learning IIb: Optimization Properties of SGD1801.02254v1 · Chiyuan Zhang, Qianli Liao, Alexander Rakhlin et al.2018 · 42 citationsarXiv
Theory II: Landscape of the Empirical Risk in Deep Learning1703.09833v2 · Qianli Liao, Tomaso Poggio2017 · 45 citationsarXiv
Why and When Can Deep -- but Not Shallow -- Networks Avoid the Curse of Dimensionality: a Review1611.00740v5 · Tomaso Poggio, Hrushikesh Mhaskar, Lorenzo Rosasco et al.2016 · 607 citationsarXiv
Compression of Deep Neural Networks for Image Instance Retrieval1701.04923v1 · Vijay Chandrasekhar, Jie Lin, Qianli Liao et al.2017 · 24 citationsarXiv
Streaming Normalization: Towards Simpler and More Biologically-plausible Normalizations for Online and Recurrent Learning1610.06160v1 · Qianli Liao, Kenji Kawaguchi, Tomaso Poggio2016 · 22 citationsarXiv
View-tolerant face recognition and Hebbian learning imply mirror-symmetric neural tuning to head orientation1606.01552v1 · Joel Z. Leibo, Qianli Liao, Winrich Freiwald et al.2016 · 57 citationsarXiv
Learning Functions: When Is Deep Better Than Shallow1603.00988v4 · Hrushikesh Mhaskar, Qianli Liao, Tomaso Poggio2016 · 105 citationsarXiv
The Invariance Hypothesis Implies Domain-Specific Regions in Visual Cortex.26496457 · Leibo, Joel Z, Liao, Qianli, Anselmi, Fabio et al.2016 · 31 citationsPLoS computational biology. 2015;11(10):e1004390
How Important is Weight Symmetry in Backpropagation?1510.05067v4 · Qianli Liao, Joel Z. Leibo, Tomaso Poggio2015 · 140 citationsarXiv
The invariance hypothesis implies domain-specific regions in visual cortex10.1101/004473v2 · Joel Z. Leibo, Qianli Liao, Fabio Anselmi et al.2014 · 31 citationsbioRxiv
Can a biologically-plausible hierarchy effectively replace face detection, alignment, and recognition pipelines?1311.4082v3 · Qianli Liao, Joel Z Leibo, Youssef Mroueh et al.2013 · 13 citationsarXiv
Career total: 70 works. 26 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.