arXiv subjects

Masanori Hirano

Publications and source records attributed to Masanori Hirano.

22 records · Page 2Linked to original sources

arXiv2023

llm-japanese-dataset v0: Construction of Japanese Chat Dataset for Large Language Models and its Methodology

This study constructed a Japanese chat dataset for tuning large language models (LLMs), which consist of about 8.4 million records. Recently, LLMs have been developed and gaining popularity. However, high-performing LLMs are usually mainly for English. There are two ways to support languages other than English by those LLMs: constructing LLMs from scratch or tuning existing models. However, in both ways, datasets are necessary parts. In this study, we focused on supporting Japanese in those LLMs and making a dataset for training or tuning LLMs in Japanese. The dataset we constructed consisted of various tasks, such as translation and knowledge tasks. In our experiment, we tuned an existing LLM using our dataset and evaluated the performance qualitatively. The results suggest that our dataset is possibly beneficial for LLMs. However, we also revealed some difficulties in constructing LLMs in languages other than English.

cs.CL↗

arXiv2023

Efficient Learning of Nested Deep Hedging using Multiple Options

Deep hedging is a framework for hedging derivatives in the presence of market frictions. In this study, we focus on the problem of hedging a given target option by using multiple options. To extend the deep hedging framework to this setting, the options used as hedging instruments also have to be priced during training. While one might use classical pricing model such as the Black-Scholes formula, ignoring frictions can offer arbitrage opportunities which are undesirable for deep hedging learning. The goal of this study is to develop a nested deep hedging method. That is, we develop a fully-deep approach of deep hedging in which the hedging instruments are also priced by deep neural networks that are aware of frictions. However, since the prices of hedging instruments have to be calculated under many different conditions, the entire learning process can be computationally intractable. To overcome this problem, we propose an efficient learning method for nested deep hedging. Our method consists of three techniques to circumvent computational intractability, each of which reduces redundant computations during training. We show through experiments that the Black-Scholes pricing of hedge instruments can admit significant arbitrage opportunities, which are not observed when the pricing is performed by deep hedging. We also demonstrate that our proposed method successfully reduces the hedging risks compared to a baseline method that does not use options as hedging instruments.

q-fin.CP↗

arXiv2022

Policy Gradient Stock GAN for Realistic Discrete Order Data Generation in Financial Markets

This study proposes a new generative adversarial network (GAN) for generating realistic orders in financial markets. In some previous works, GANs for financial markets generated fake orders in continuous spaces because of GAN architectures' learning limitations. However, in reality, the orders are discrete, such as order prices, which has minimum order price unit, or order types. Thus, we change the generation method to place the generated fake orders into discrete spaces in this study. Because this change disabled the ordinary GAN learning algorithm, this study employed a policy gradient, frequently used in reinforcement learning, for the learning algorithm. Through our experiments, we show that our proposed model outperforms previous models in generated order distribution. As an additional benefit of introducing the policy gradient, the entropy of the generated policy can be used to check GAN's learning status. In the future, higher performance GANs, better evaluation methods, or the applications of our GANs can be addressed.

cs.LG↗

arXiv2012

Potential antiferromagnetic fluctuations in hole-doped iron-pnictide superconductor Ba_{1-x}K_{x}Fe_{2}As_{2} studied by ^{75}As nuclear magnetic

We have performed ^{75}As nuclear magnetic resonance (NMR) and nuclear quadrupole resonance (NQR) on single crystalline Ba_{1-x}K_{x}Fe_{2}As_{2} for x = 0.27-1. ^{75}As nuclear quadruple resonance frequency (ν_{Q}) increases linearly with increasing x. The Knight shift K in normal state shows Pauli paramagnetic behavior with slight temperature T dependence. The value of K increases gradually with increasing x. By contrast, nuclear spin- lattice relaxation rate 1/T_{1} in normal state has a large T-dependence, which indicates existence of large antiferomagnetic (AF) spin fluctuations for all x. The T-dependence of 1/T_{1} shows a gap-like behavior below approximately 100 K for 0.6 < x < 0.9. These behaviors are well explained by the change of band structure with expansion of hole Fermi surfaces and shrink and disappearance of electron Fermi surfaces at Brillouin zone (BZ) with increasing x. The anisotropy of 1/T_{1}, represented by a ratio of 1/T_{1ab} to 1/T_{1c}, is always larger than 1 for all x, which indicates that the stripe-type AF fluctuations is dominant in this system. The K in superconducting (SC) state decreases, which corresponds to appearance of spin-singlet superconductivity. The T dependence of 1/T_{1} in SC state indicates multiple-SC-gap feature. A simple two gap model analysis shows that the larger superconducting gap gradually decreases with increasing x from 0.27 to 1 and smaller gap decreases rapidly and nearly vanishes for x > 0.6 where the electron pockets in BZ disappear.

cond-mat.supr-con↗