Lech Szymanski

Publication Activity (10 Years)

Years Active: 2005-2024
Publications (10 Years): 27

Top Topics

Hopfield Network

Document Search

Reinforcement Learning

Hypothesis Space

Top Venues

IEEE Trans. Neural Networks Learn. Syst.

Publications

Hayden McAlister, Anthony Robins, Lech Szymanski
Improved Robustness and Hyperparameter Selection in Modern Hopfield Networks. CoRR (2024)
Hayden McAlister, Anthony Robins, Lech Szymanski
Prototype Analysis in Hopfield Networks with Hebbian Learning. CoRR (2024)
Haitao Xu, Lech Szymanski, Brendan McCane
VASE: Variational Assorted Surprise Exploration for Reinforcement Learning. IEEE Trans. Neural Networks Learn. Syst. 34 (3) (2023)
Lech Szymanski, Brendan McCane, Craig Atkinson
Conceptual complexity of neural networks. Neurocomputing 469 (2022)
Craig Atkinson, Brendan McCane, Lech Szymanski, Anthony V. Robins
Pseudo-rehearsal: Achieving deep reinforcement learning without catastrophic forgetting. Neurocomputing 428 (2021)
Lech Szymanski, Brendan McCane, Craig Atkinson
Conceptual capacity and effective complexity of neural networks. CoRR (2021)
Lech Szymanski, Michael Lee
Coarse facial feature detection in sheep. IVCNZ (2021)
Yerren van Sint Annaland, Lech Szymanski, Steven Mills
Predicting Cherry Quality Using Siamese Networks. IVCNZ (2020)
Lech Szymanski, Michael Lee
Deep Sheep: kinship assignment in livestock from facial images. IVCNZ (2020)
Haitao Xu, Brendan McCane, Lech Szymanski, Craig Atkinson
MIME: Mutual Information Minimisation Exploration. CoRR (2020)
Haitao Xu, Brendan McCane, Lech Szymanski
VASE: Variational Assorted Surprise Exploration for Reinforcement Learning. CoRR (2019)
Lech Szymanski, Brendan McCane, Craig Atkinson
Switched linear projections and inactive state sensitivity for deep neural network interpretability. CoRR (2019)
Chris Gorman, Lech Szymanski, Anthony V. Robins, Alistair Knott
A Convolutional Self-organizing Map for Visual Category Learning. CogSci (2019)
Craig Atkinson, Brendan McCane, Lech Szymanski, Anthony V. Robins
GRIm-RePR: Prioritising Generating Important Features for Pseudo-Rehearsal. CoRR (2019)
Hannah Clark-Younger, Steven Mills, Lech Szymanski
Stacked Hourglass CNN for Handwritten Character Location. IVCNZ (2018)
Brendan McCane, Lech Szymanski
Efficiency of deep networks for radially symmetric functions. Neurocomputing 313 (2018)
Craig Atkinson, Brendan McCane, Lech Szymanski, Anthony V. Robins
Pseudo-Rehearsal: Achieving Deep Reinforcement Learning without Catastrophic Forgetting. CoRR (2018)
Lech Szymanski, Brendan McCane, Michael Albert
The effect of the choice of neural network depth and breadth on the size of its hypothesis space. CoRR (2018)
Haitao Xu, Brendan McCane, Lech Szymanski
Twin Bounded Large Margin Distribution Machine. Australasian Conference on Artificial Intelligence (2018)
Craig Atkinson, Brendan McCane, Lech Szymanski, Anthony V. Robins
Pseudo-Recursal: Solving the Catastrophic Forgetting Problem in Deep Neural Networks. CoRR (2018)
Brendan McCane, Lech Szymanski
Some Approximation Bounds for Deep Networks. CoRR (2018)
Lech Szymanski, Brendan McCane, Wei Gao, Zhi-Hua Zhou
Effects of the optimisation of the margin distribution on generalisation in deep architectures. CoRR (2017)
Brendan McCane, Lech Szymanski
Deep Radial Kernel Networks: Approximating Radially Symmetric Functions with Deep Networks. CoRR (2017)
Craig Atkinson, Brendan McCane, Lech Szymanski
Increasing the accuracy of convolutional neural networks with progressive reinitialisation. IVCNZ (2017)
Lech Szymanski, Steven Mills
CNN for historic handwritten document search. IVCNZ (2017)
Brendan McCane, Lech Szymanski
Deep networks are efficient for circular manifolds. ICPR (2016)
Xiping Fu, Brendan McCane, Steven Mills, Michael Albert, Lech Szymanski
Auto-JacoBin: Auto-encoder Jacobian Binary Hashing. CoRR (2016)
Steven Mills, Lech Szymanski, Reuben Johnson
Hierarchical Structure from Motion from Endoscopic Video. IVCNZ (2014)
Lech Szymanski, Brendan McCane
Deep Networks are Effective Encoders of Periodicity. IEEE Trans. Neural Networks Learn. Syst. 25 (10) (2014)
Lech Szymanski, Brendan McCane
Learning in deep architectures with folding transformations. IJCNN (2013)
Shawn Martin, Lech Szymanski
Singularity resolution for dimension reduction. IVCNZ (2013)
Lech Szymanski, Brendan McCane
Push-pull separability objective for supervised layer-wise training of neural networks. IJCNN (2012)
Lech Szymanski, Brendan McCane
Deep, super-narrow neural network is a universal classifier. IJCNN (2012)
Lech Szymanski, Martin Bouchard
Comb filter decomposition for robust ASR. INTERSPEECH (2005)