34 citations · 77 across the 22 of their papers we have counts for
23 papers
Structured Pruning of Large Language Models via Power Transformation and Sign-Preserving Score Aggregation with Adaptive Feature Retention
Ryota Kobayashi, Tsubasa Hirakawa, Takayoshi Yamashita +4
This paper proposes an improved structured pruning method for large language models (LLMs) that addresses key challenges in adapting Adaptive Feature Retention (AFR), an unstructur…
P2GS: Physical Prior-guided Gaussian Splatting for Photometrically Consistent Urban Reconstruction
Kota Shimomura, Hidehisa Arai, Tsubasa Takahashi +2
3D Gaussian Splatting (3DGS) has recently emerged as a powerful explicit representation enabling fast, high-fidelity rendering, making it a promising foundation for closed-loop sim…
Mix-Geneformer: Unified Representation Learning for Human and Mouse scRNA-seq Data
Yuki Nishio, Takayoshi Yamashita, Keita Ito +2
Single-cell RNA sequencing (scRNA-seq) enables single-cell transcriptomic profiling, revealing cellular heterogeneity and rare populations. Recent deep learning models like Genefor…
DeBiFormer: Vision Transformer with Deformable Agent Bi-level Routing Attention
Nguyen Huu Bao Long, Chenyu Zhang, Yuzhi Shi +4
Vision Transformers with various attention modules have demonstrated superior performance on vision tasks. While using sparsity-adaptive attention, such as in DAT, has yielded stro…
Nearest Neighbor Future Captioning: Generating Descriptions for Possible Collisions in Object Placement Tasks
Takumi Komatsu, Motonari Kambara, Shumpei Hatanaka +5
Domestic service robots (DSRs) that support people in everyday environments have been widely investigated. However, their ability to predict and describe future risks resulting fro…
Layer-Wise Relevance Propagation with Conservation Property for ResNet
Seitaro Otsuki, Tsumugi Iida, Félix Doublet +4
The transparent formulation of explanation methods is essential for elucidating the predictions of neural networks, which are typically black-box models. Layer-wise Relevance Propa…
Action Q-Transformer: Visual Explanation in Deep Reinforcement Learning with Encoder-Decoder Model using Action Query
Hidenori Itaya, Tsubasa Hirakawa, Takayoshi Yamashita +2
The excellent performance of Transformer in supervised learning has led to growing interest in its potential application to deep reinforcement learning (DRL) to achieve high perfor…
Learning from AI: An Interactive Learning Method Using a DNN Model Incorporating Expert Knowledge as a Teacher
Kohei Hattori, Tsubasa Hirakawa, Takayoshi Yamashita +1
Visual explanation is an approach for visualizing the grounds of judgment by deep learning, and it is possible to visually interpret the grounds of a judgment for a certain input b…
Masking and Mixing Adversarial Training
Hiroki Adachi, Tsubasa Hirakawa, Takayoshi Yamashita +3
While convolutional neural networks (CNNs) have achieved excellent performances in various computer vision tasks, they often misclassify with malicious samples, a.k.a. adversarial…
Data Augmentation by Selecting Mixed Classes Considering Distance Between Classes
Shungo Fujii, Yasunori Ishii, Kazuki Kozuka +3
Data augmentation is an essential technique for improving recognition accuracy in object recognition using deep learning. Methods that generate mixed data from multiple data sets,…
Visual Explanation of Deep Q-Network for Robot Navigation by Fine-tuning Attention Branch
Yuya Maruyama, Hiroshi Fukui, Tsubasa Hirakawa +3
Robot navigation with deep reinforcement learning (RL) achieves higher performance and performs well under complex environment. Meanwhile, the interpretation of the decision-making…
Object-ABN: Learning to Generate Sharp Attention Maps for Action Recognition
Tomoya Nitta, Tsubasa Hirakawa, Hironobu Fujiyoshi +1
In this paper we propose an extension of the Attention Branch Network (ABN) by using instance segmentation for generating sharper attention maps for action recognition. Methods for…
ST-ABN: Visual Explanation Taking into Account Spatio-temporal Information for Video Recognition
Masahiro Mitsuhara, Tsubasa Hirakawa, Takayoshi Yamashita +1
It is difficult for people to interpret the decision-making in the inference process of deep neural networks. Visual explanation is one method for interpreting the decision-making…
Deep Ensemble Collaborative Learning by using Knowledge-transfer Graph for Fine-grained Object Classification
Naoki Okamoto, Soma Minami, Tsubasa Hirakawa +2
Mutual learning, in which multiple networks learn by sharing their knowledge, improves the performance of each network. However, the performance of ensembles of networks that have…
Visual Explanation using Attention Mechanism in Actor-Critic-based Deep Reinforcement Learning
Hidenori Itaya, Tsubasa Hirakawa, Takayoshi Yamashita +2
Deep reinforcement learning (DRL) has great potential for acquiring the optimal action in complex environments such as games and robot control. However, it is difficult to analyze…
Predicting and Attending to Damaging Collisions for Placing Everyday Objects in Photo-Realistic Simulations
Aly Magassouba, Komei Sugiura, Angelica Nakayama +4
Placing objects is a fundamental task for domestic service robots (DSRs). Thus, inferring the collision-risk before a placing motion is crucial for achieving the requested task. Th…
Alleviating the Burden of Labeling: Sentence Generation by Attention Branch Encoder-Decoder Network
Tadashi Ogura, Aly Magassouba, Komei Sugiura +4
Domestic service robots (DSRs) are a promising solution to the shortage of home care workers. However, one of the main limitations of DSRs is their inability to interact naturally…
Knowledge Transfer Graph for Deep Collaborative Learning
Soma Minami, Tsubasa Hirakawa, Takayoshi Yamashita +1
Knowledge transfer among multiple networks using their outputs or intermediate activations have evolved through extensive manual design from a simple teacher-student approach (know…
Embedding Human Knowledge into Deep Neural Network via Attention Map
Masahiro Mitsuhara, Hiroshi Fukui, Yusuke Sakashita +4
In this work, we aim to realize a method for embedding human knowledge into deep neural networks. While the conventional method to embed human knowledge has been applied for non-de…
Attention Branch Network: Learning of Attention Mechanism for Visual Explanation
Hiroshi Fukui, Tsubasa Hirakawa, Takayoshi Yamashita +1
Visual explanation enables human to understand the decision making of Deep Convolutional Neural Network (CNN), but it is insufficient to contribute the performance improvement. In…
Survey on Vision-based Path Prediction
Tsubasa Hirakawa, Takayoshi Yamashita, Toru Tamaki +1
Path prediction is a fundamental task for estimating how pedestrians or vehicles are going to move in a scene. Because path prediction as a task of computer vision uses video as in…
Denoising random forests
Masaya Hibino, Akisato Kimura, Takayoshi Yamashita +2
This paper proposes a novel type of random forests called a denoising random forests that are robust against noises contained in test samples. Such noise-corrupted samples cause se…
Binary-decomposed DCNN for accelerating computation and compressing model without retraining
Ryuji Kamiya, Takayoshi Yamashita, Mitsuru Ambai +3
Recent trends show recognition accuracy increasing even more profoundly. Inference process of Deep Convolutional Neural Networks (DCNN) has a large number of parameters, requires a…