no code implementations • 16 Apr 2024 • Masanori Hirano, Kentaro Imajo
After continual pre-training using the datasets and the base model, the tuned model performed better than the original model on the Japanese financial benchmarks.
no code implementations • 15 Apr 2024 • Masanori Hirano
This study proposes a new approach using artificial market simulations for underlying asset simulations in deep hedging.
1 code implementation • 22 Mar 2024 • Masanori Hirano
With the recent development of large language models (LLMs), models that focus on certain domains and languages have been discussed for their necessity.
no code implementations • 13 Nov 2023 • Rawin Assabumrungrat, Kentaro Minami, Masanori Hirano
Through comparative experiments, we assessed the empirical performance of these solvers in high-dimensional contexts.
1 code implementation • 19 Sep 2023 • Masanori Hirano, Ryosuke Takata, Kiyoshi Izumi
This paper presents a new artificial market simulation platform, PAMS: Platform for Artificial Market Simulations.
1 code implementation • 7 Sep 2023 • Masahiro Suzuki, Masanori Hirano, Hiroki Sakaji
We performed Low-Rank Adaptation (LoRA) tuning on both Japanese and English existing models using our instruction dataset.
no code implementations • 25 Jul 2023 • Masanori Hirano, Kentaro Minami, Kentaro Imajo
In this framework, a hedger and a generator, which respectively model the underlying asset process and the underlying asset process, are trained in an adversarial manner.
1 code implementation • 22 May 2023 • Masanori Hirano, Masahiro Suzuki, Hiroki Sakaji
There are two ways to support languages other than English by those LLMs: constructing LLMs from scratch or tuning existing models.
no code implementations • 20 May 2023 • Masanori Hirano, Kentaro Imajo, Kentaro Minami, Takuya Shimada
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.
no code implementations • 28 Apr 2022 • Masanori Hirano, Hiroki Sakaji, Kiyoshi Izumi
In some previous works, GANs for financial markets generated fake orders in continuous spaces because of GAN architectures' learning limitations.