Tomotake Sasaki

发表

When and how convolutional neural networks generalize to out-of-distribution category–viewpoint combinations

F. Durand, X. Boix, H. Pfister, 2022, Nat. Mach. Intell..

Recurrent connections facilitate symmetry perception in deep networks

X. Boix, Tomotake Sasaki, Shobhita Sundaram, 2022, Scientific Reports.

Rate-distortion optimization guided autoencoder for isometric embedding in Euclidean latent space

Keizo Kato, Akira Nakagawa, Tomotake Sasaki, 2019, ICML.

Do Neural Networks for Segmentation Understand Insideness?

Xavier Boix, Tomotake Sasaki, Andrew Francl, 2020, Neural Computation.

Deephys: Deep Electrophysiology, Debugging Neural Networks under Distribution Shifts

X. Boix, Anirban Sarkar, I. Mason, 2023, ArXiv.

Small in-distribution changes in 3D perspective and lighting fool both CNNs and Transformers pdf

Hanspeter Pfister, Tzu-Mao Li, Xavier Boix, 2021, ArXiv.

Adversarial examples within the training distribution: A widespread challenge

X. Boix, H. Pfister, Tzu-Mao Li, 2021, 2106.16198.

HICO-DET-SG and V-COCO-SG: New Data Splits to Evaluate Systematic Generalization in Human-Object Interaction Detection

H. Akima, Kenta Takemoto, Tomotake Sasaki, 2023, arXiv.org.

When and how CNNs generalize to out-of-distribution category-viewpoint combinations

F. Durand, X. Boix, H. Pfister, 2020, 2007.08032.

Transformer Module Networks for Systematic Generalization in Visual Question Answering

X. Boix, Kenta Takemoto, Tomotake Sasaki, 2022, ArXiv.

How Modular Should Neural Module Networks Be for Systematic Generalization? pdf

Vanessa D'Amario, Xavier Boix, Tomotake Sasaki, 2021, NeurIPS.

Can Deep Neural Networks Do Image Segmentation by Understanding Insideness?

T. Poggio, X. Boix, Vilim Štih, 2018 .

HICO-DET-SG and V-COCO-SG: New Data Splits for Evaluating the Systematic Generalization Performance of Human-Object Interaction Detection Models

H. Akima, Kenta Takemoto, Tomotake Sasaki, 2023, ArXiv.

Modularity Trumps Invariance for Compositional Robustness

X. Boix, Anirban Sarkar, I. Mason, 2023, ArXiv.

Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations: late-stopping, tuning batch normalization and invariance loss pdf

Hanspeter Pfister, Xavier Boix, Spandan Madan, 2021, ArXiv.

To Which Out-Of-Distribution Object Orientations Are DNNs Capable of Generalizing? pdf

Hanspeter Pfister, Pawan Sinha, Xavier Boix, 2021, ArXiv.

Emergent Neural Network Mechanisms for Generalization to Objects in Novel Orientations

X. Boix, H. Pfister, Daniel Harari, 2021, 2109.13445.

On the Capability of Neural Networks to Generalize to Unseen Category-Pose Combinations pdf

Hanspeter Pfister, Xavier Boix, Tomotake Sasaki, 2020, ArXiv.

Symmetry Perception by Deep Networks: Inadequacy of Feed-Forward Architectures and Improvements with Recurrent Connections pdf

Xavier Boix, Tomotake Sasaki, Shobhita Sundaram, 2021, ArXiv.

Three approaches to facilitate invariant neurons and generalization to out-of-distribution orientations and illuminations

Kanata Suzuki, X. Boix, H. Pfister, 2022, Neural Networks.

The Foes of Neural Network's Data Efficiency Among Unnecessary Input Dimensions pdf

Xavier Boix, Sanjana Srivastava, Vanessa D'Amario, 2021, ArXiv.

The Data Efficiency of Deep Learning Is Degraded by Unnecessary Input Dimensions

X. Boix, Tomotake Sasaki, Vanessa D'Amario, 2022, Frontiers in Computational Neuroscience.

Annotation Cost Reduction of Stream-based Active Learning by Automated Weak Labeling using a Robot Arm

Kanata Suzuki, Takashi Katoh, Tomotake Sasaki, 2021, 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).