Zobrazeno 1 - 10
of 85
pro vyhledávání: '"Sugiyama, Mahito"'
Autor:
Enouen, James, Sugiyama, Mahito
The log-linear model has received a significant amount of theoretical attention in previous decades and remains the fundamental tool used for learning probability distributions over discrete variables. Despite its large popularity in statistical mech
Externí odkaz:
http://arxiv.org/abs/2410.11964
Autor:
Hu, Pingbang, Sugiyama, Mahito
We propose a novel and interpretable data augmentation method based on energy-based modeling and principles from information geometry. Unlike black-box generative models, which rely on deep neural networks, our approach replaces these non-interpretab
Externí odkaz:
http://arxiv.org/abs/2410.00718
Autor:
Kanoh, Ryuichi, Sugiyama, Mahito
Linear Mode Connectivity (LMC) refers to the phenomenon that performance remains consistent for linearly interpolated models in the parameter space. For independently optimized model pairs from different random initializations, achieving LMC is consi
Externí odkaz:
http://arxiv.org/abs/2405.14596
Autor:
Cheema, Prasad, Sugiyama, Mahito
Data augmentation is an area of research which has seen active development in many machine learning fields, such as in image-based learning models, reinforcement learning for self driving vehicles, and general noise injection for point cloud data. Ho
Externí odkaz:
http://arxiv.org/abs/2402.19287
Autor:
Yamada, Masatsugu, Sugiyama, Mahito
Designing molecular structures with desired chemical properties is an essential task in drug discovery and material design. However, finding molecules with the optimized desired properties is still a challenging task due to combinatorial explosion of
Externí odkaz:
http://arxiv.org/abs/2302.00587
We present an alternative approach to decompose non-negative tensors, called many-body approximation. Traditional decomposition methods assume low-rankness in the representation, resulting in difficulties in global optimization and target rank select
Externí odkaz:
http://arxiv.org/abs/2209.15338
Autor:
Kanoh, Ryuichi, Sugiyama, Mahito
A soft tree is an actively studied variant of a decision tree that updates splitting rules using the gradient method. Although soft trees can take various architectures, their impact is not theoretically well known. In this paper, we formulate and an
Externí odkaz:
http://arxiv.org/abs/2205.12904
Autor:
Ghalamkari, Kazu, Sugiyama, Mahito
We propose a fast non-gradient-based method of rank-1 non-negative matrix factorization (NMF) for missing data, called A1GM, that minimizes the KL divergence from an input matrix to the reconstructed rank-1 matrix. Our method is based on our new find
Externí odkaz:
http://arxiv.org/abs/2110.12595
Autor:
Kanoh, Ryuichi, Sugiyama, Mahito
In practical situations, the tree ensemble is one of the most popular models along with neural networks. A soft tree is a variant of a decision tree. Instead of using a greedy method for searching splitting rules, the soft tree is trained using a gra
Externí odkaz:
http://arxiv.org/abs/2109.04983
Autor:
Kanoh, Ryuichi, Sugiyama, Mahito
A multiplicative constant scaling factor is often applied to the model output to adjust the dynamics of neural network parameters. This has been used as one of the key interventions in an empirical study of lazy and active behavior. However, we show
Externí odkaz:
http://arxiv.org/abs/2103.03466