Search Results for author: Tomotake Sasaki

Found 20 papers, 12 papers with code

D3: Data Diversity Design for Systematic Generalization in Visual Question Answering

1 code implementation15 Sep 2023 Amir Rahimi, Vanessa D'Amario, Moyuru Yamada, Kentaro Takemoto, Tomotake Sasaki, Xavier Boix

We demonstrate that this result is independent of the similarity between the training and testing data and applies to well-known families of neural network architectures for VQA (i. e. monolithic architectures and neural module networks).

Question Answering Systematic Generalization +1

Modularity Trumps Invariance for Compositional Robustness

1 code implementation15 Jun 2023 Ian Mason, Anirban Sarkar, Tomotake Sasaki, Xavier Boix

In this work we develop a compositional image classification task where, given a few elemental corruptions, models are asked to generalize to compositions of these corruptions.

Domain Generalization Image Classification

Safe Exploration Method for Reinforcement Learning under Existence of Disturbance

1 code implementation30 Sep 2022 Yoshihiro Okawa, Tomotake Sasaki, Hitoshi Yanami, Toru Namerikawa

We define the safety during learning as satisfaction of the constraint conditions explicitly defined in terms of the state and propose a safe exploration method that uses partial prior knowledge of a controlled object and disturbance.

reinforcement-learning Reinforcement Learning (RL) +1

Transformer Module Networks for Systematic Generalization in Visual Question Answering

1 code implementation27 Jan 2022 Moyuru Yamada, Vanessa D'Amario, Kentaro Takemoto, Xavier Boix, Tomotake Sasaki

We reveal that Neural Module Networks (NMNs), i. e., question-specific compositions of modules that tackle a sub-task, achieve better or similar systematic generalization performance than the conventional Transformers, even though NMNs' modules are CNN-based.

Question Answering Systematic Generalization +1

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

1 code implementation8 Dec 2021 Shobhita Sundaram, Darius Sinha, Matthew Groth, Tomotake Sasaki, Xavier Boix

Symmetry is omnipresent in nature and perceived by the visual system of many species, as it facilitates detecting ecologically important classes of objects in our environment.

Three approaches to facilitate DNN generalization to objects in out-of-distribution orientations and illuminations

1 code implementation30 Oct 2021 Akira Sakai, Taro Sunagawa, Spandan Madan, Kanata Suzuki, Takashi Katoh, Hiromichi Kobashi, Hanspeter Pfister, Pawan Sinha, Xavier Boix, Tomotake Sasaki

While humans have a remarkable capability of recognizing objects in out-of-distribution (OoD) orientations and illuminations, Deep Neural Networks (DNNs) severely suffer in this case, even when large amounts of training examples are available.

Emergent Neural Network Mechanisms for Generalization to Objects in Novel Orientations

no code implementations28 Sep 2021 Avi Cooper, Xavier Boix, Daniel Harari, Spandan Madan, Hanspeter Pfister, Tomotake Sasaki, Pawan Sinha

The capability of Deep Neural Networks (DNNs) to recognize objects in orientations outside the distribution of the training data is not well understood.

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

no code implementations13 Jul 2021 Vanessa D'Amario, Sanjana Srivastava, Tomotake Sasaki, Xavier Boix

Datasets often contain input dimensions that are unnecessary to predict the output label, e. g. background in object recognition, which lead to more trainable parameters.

Object Recognition

Adversarial examples within the training distribution: A widespread challenge

1 code implementation30 Jun 2021 Spandan Madan, Tomotake Sasaki, Hanspeter Pfister, Tzu-Mao Li, Xavier Boix

This result provides evidence supporting theories attributing adversarial examples to the proximity of data to ground-truth class boundaries, and calls into question other theories which do not account for this more stringent definition of adversarial attacks.

Object Recognition Open-Ended Question Answering

Two-step reinforcement learning for model-free redesign of nonlinear optimal regulator

1 code implementation5 Mar 2021 Mei Minami, Yuka Masumoto, Yoshihiro Okawa, Tomotake Sasaki, Yutaka Hori

To overcome this limitation, we propose a model-free two-step design approach that improves the transient learning performance of RL in an optimal regulator redesign problem for unknown nonlinear systems.

Offline RL reinforcement-learning +1

Automatic Exploration Process Adjustment for Safe Reinforcement Learning with Joint Chance Constraint Satisfaction

no code implementations5 Mar 2021 Yoshihiro Okawa, Tomotake Sasaki, Hidenao Iwane

In reinforcement learning (RL) algorithms, exploratory control inputs are used during learning to acquire knowledge for decision making and control, while the true dynamics of a controlled object is unknown.

Decision Making Object +3

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

no code implementations1 Jan 2021 Vanessa D'Amario, Sanjana Srivastava, Tomotake Sasaki, Xavier Boix

In this paper, we investigate the impact of unnecessary input dimensions on one of the central issues of machine learning: the number of training examples needed to achieve high generalization performance, which we refer to as the network's data efficiency.

Foveation Image Classification +3

On the Capability of CNNs to Generalize to Unseen Category-Viewpoint Combinations

no code implementations1 Jan 2021 Spandan Madan, Timothy Henry, Jamell Arthur Dozier, Helen Ho, Nishchal Bhandari, Tomotake Sasaki, Fredo Durand, Hanspeter Pfister, Xavier Boix

We find that learning category and viewpoint in separate networks compared to a shared one leads to an increase in selectivity and invariance, as separate networks are not forced to preserve information about both category and viewpoint.

Object Recognition Viewpoint Estimation

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

2 code implementations15 Jul 2020 Spandan Madan, Timothy Henry, Jamell Dozier, Helen Ho, Nishchal Bhandari, Tomotake Sasaki, Frédo Durand, Hanspeter Pfister, Xavier Boix

In this paper, we investigate when and how such OOD generalization may be possible by evaluating CNNs trained to classify both object category and 3D viewpoint on OOD combinations, and identifying the neural mechanisms that facilitate such OOD generalization.

Object Recognition Viewpoint Estimation

Rate-Distortion Optimization Guided Autoencoder for Isometric Embedding in Euclidean Latent Space

no code implementations ICML 2020 Keizo Kato, Jing Zhou, Tomotake Sasaki, Akira Nakagawa

We show our method has the following properties: (i) the Jacobian matrix between the input space and a Euclidean latent space forms a constantlyscaled orthonormal system and enables isometric data embedding; (ii) the relation of PDFs in both spaces can become tractable one such as proportional relation.

Relation Unsupervised Anomaly Detection

Do Deep Neural Networks for Segmentation Understand Insideness?

no code implementations25 Sep 2019 Kimberly M Villalobos, Vilim Stih, Amineh Ahmadinejad, Jamell Dozier, Andrew Francl, Frederico Azevedo, Tomotake Sasaki, Xavier Boix

At the heart of image segmentation lies the problem of determining whether a pixel is inside or outside a region, which we denote as the "insideness" problem.

Image Segmentation Segmentation +1

Cannot find the paper you are looking for? You can Submit a new open access paper.