no code implementations • 16 Jul 2021 • Minchao Wu, Takeshi Tsukada, Hiroshi Unno, Taro Sekiyama, Kohei Suenaga
As a concrete example, we implement the approach on top of PCSat, which is an invariant synthesizer based on template-based CEGIS.
1 code implementation • 17 Mar 2021 • Naoki Kobayashi, Taro Sekiyama, Issei Sato, Hiroshi Unno
Another application is to a new program development framework called oracle-based programming, which is a neural-network-guided variation of Solar-Lezama's program synthesis by sketching.
1 code implementation • 18 Mar 2020 • Taro Sekiyama, Takeshi Tsukada, Atsushi Igarashi
We propose signature restriction, a new notion to restrict the type signatures of operations, and show that signature restriction is sufficient to ensure type safety of an effectful language equipped with unrestricted polymorphic type assignment.
Programming Languages
no code implementations • 18 Oct 2019 • Taro Sekiyama, Atsushi Igarashi
Key ingredients in this work are the dynamic row type, which represents a statically unknown part of a row, and consistency for row types, which allows injecting static row types into the dynamic row type and, conversely, projecting the dynamic row type to any static row type.
Programming Languages
no code implementations • 5 Apr 2019 • Takamasa Okudono, Masaki Waga, Taro Sekiyama, Ichiro Hasuo
We present a method to extract a weighted finite automaton (WFA) from a recurrent neural network (RNN).
no code implementations • 18 Nov 2018 • Taro Sekiyama, Atsushi Igarashi
Algebraic effects and handlers are a powerful abstraction mechanism to represent and implement control effects.
Programming Languages
1 code implementation • 30 Oct 2018 • Yusuke Miyazaki, Taro Sekiyama, Atsushi Igarashi
The DTI-based semantics not only avoids the divergence described above but also is sound and complete with respect to the semantics of fully instantiated terms in the following sense: if the evaluation of a term succeeds (i. e., terminates with a value) in the DTI-based semantics, then there is a fully instantiated version of the term that also succeeds in the explicitly typed blame calculus and vice versa.
Programming Languages Logic in Computer Science
1 code implementation • 30 May 2018 • Taro Sekiyama, Kohei Suenaga
As an implementation of the estimator, we propose a proposition-to-proof architecture, which is a DNN tailored to the automated proof synthesis problem.
no code implementations • 26 Apr 2018 • Taro Sekiyama, Takashi Imamichi, Haruki Imai, Rudy Raymond
We address this challenge by developing a novel profile-guided memory optimization to efficiently and quickly allocate memory blocks during the propagation in DNNs.
no code implementations • 19 Nov 2017 • Masaharu Sakamoto, Hiroki Nakano, Kun Zhao, Taro Sekiyama
In this paper, we propose Fusion classifier in conjunction with the cascaded convolutional neural network models.
no code implementations • ICML 2017 • Takayuki Osogami, Hiroshi Kajino, Taro Sekiyama
Hidden units can play essential roles in modeling time-series having long-term dependency or on-linearity but make it difficult to learn associated parameters.
no code implementations • 20 Jun 2017 • Taro Sekiyama, Akifumi Imanishi, Kohei Suenaga
Inspired by the recent evolution of deep neural networks (DNNs) in machine learning, we explore their application to PL-related topics.
no code implementations • 1 Mar 2017 • Masaharu Sakamoto, Hiroki Nakano, Kun Zhao, Taro Sekiyama
Lung nodule classification is a class imbalanced problem because nodules are found with much lower frequency than non-nodules.