Category Archives: 未分類

A Paper Accepted for AAAI 2023

Our paper proposing a method for off-line deep reinforcement learning methods from mixed-behavior episode data obtained from several different policies was accepted to AAAI 2023, a top conference in the field of AI.
Guoxi Zhang, Hisashi Kashima.
Behavior Estimation from Multi-Source Data for Offline Reinforcement Learning.
In Proceedings of the 37th AAAI Conference on Artificial Intelligence (AAAI), 2022.

A Paper Accepted for IEEE BigData 2022

Our paper proposing a variational inference method for factorization machines and applying it to recommendation systems was accepted to IEEE BigData 2022.
Jill-Jênn Vie, Tomas Rigaux, Hisashi Kashima.
Variational Factorization Machines for Preference Elicitation in Large-Scale Recommender Systems.
In Proceedings of the 2022 IEEE International Conference on Big Data (BigData), 2022.

JSAI Annual Conference Award at JSAI 2022

Our paper, “Anisotropic Estimation of Station Market Areas by Supervised Learning Using Geospatial Information and IC Commuter Pass Data“, is selected as JSAI Annual Conference Award at The 36th Annual Conference of the Japanese Society for Artificial Intelligence, 2022.
Yohei Kodama, Yuki Akeyama, Yusuke Miyazaki, and Koh Takeuchi.
Anisotropic Estimation of Station Market Areas by Supervised Learning Using Geospatial Information and IC Commuter Pass Data
Proceedings of JSAI 2022
List of JSAI Annual Conference Award

Best Poster Award at ACM SIGSPATIAL 2022

Our paper, “Estimating counterfactual treatment outcomes over time in multi-vehicle simulation”, was selected as the best poster award at 30th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL 2022).

Keisuke Fujii (Nagoya University), Koh Takeuchi (Kyoto University), Atsushi Kuribayashi (Nagoya University), Naoya Takeishi (HES-SO), Yoshinobu Kawahara (Kyushu University), and Kazuya Takeda (Nagoya University).
Estimating counterfactual treatment outcomes over time in multi-vehicle simulation
(PaperVideo)

30th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (ACM SIGSPATIAL 2022) 
Tuesday November 1 – Friday November 4, 2022 Seattle, Washington, USA

Two Papers Accepted for Machine Learning Journal

A paper proposing a method for estimating intervention effects while estimating confounding variables when they are not observed, and a paper proposing a method for estimating state importance in reinforcement learning from pairwise comparisons of episodes have been accepted for publication in the Machine Learning Journal (ACML Journal Track).
Shonosuke Harada, Hisashi Kashima.
InfoCEVAE: Treatment Effect Estimation with Hidden Confounding Variables Matching.
Machine Learning, 2022.
Guoxi Zhang, Hisashi Kashima.
Learning State Importance for Preference-based Reinforcement Learning.
Machine Learning, 2022.


A Paper Accepted for DSAA 2022

Our paper proposing a novel ranking method using queries comparing rank differences of two pairs of items was accepted for the 9th IEEE International Conference on Data Science and Advanced Analytics (DSAA).
Guoxi Zhang, Jiyi Li, Hisashi Kashima.
Improving Pairwise Rank Aggregation via Querying for Rank Difference.
In Proceedings of the 9th IEEE International Conference on Data Science and Advanced Analytics (DSAA), 2022.

Three Papers Accepted for CIKM 2022

Three papers were accepted for Conference on Information and Knowledge Management (CIKM):
Ryoma Sato.
Towards Principled User-side Recommender Systems.
In Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM), 2022.
# Demonstrated the theoretical feasibility of a user-side recommendation system and proposed a user-side recommendation algorithm with desirable properties

Ryoma Sato.
CLEAR: A Fully User-side Image Search System.
In Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM), 2022.
# A demonstration working in a browser that performs image searches based on a score function set by a user.

Ryoma Sato, Makoto Yamada, Hisashi Kashima.
Twin Papers: A Simple Framework of Causal Inference for Citations via Coupling.
In Proceedings of the 31st ACM International Conference on Information and Knowledge Management (CIKM), 2022.
# Proposed a method for analyzing research impact using causal inference based on "twin papers".

A Paper Accepted for Data Mining and Knowledge Discovery (DAMI)

Our paper proposing learning a fair predictor based on the notion of “causal pathways” was accepted for Data Mining and Knowledge Discovery (DAMI) journal:
Yoichi Chikahara, Shinsaku Sakaue, Akinori Fujino, Hisashi Kashima.
Making Individually Fair Predictions with Causal Pathways.
Data Mining and Knowledge Discovery (DAMI), 2022.

Two Papers Accepted for ECML PKDD 2022

Two papers were accepted for ECML PKDD, a premier conference on machine learning and data mining:
Guoxi Zhang, Hisashi Kashima.
Batch Reinforcement Learning from Crowds.
In Proceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), 2022.

Mathis Petrovich, Chao Liang, Ryoma Sato, Yanbin Liu, Yao-Hung Hubert Tsai, Linchao Zhu, Yi Yang, Ruslan Salakhutdinov, Makoto Yamada.
Feature Robust Optimal Transport for High-dimensional Data.
In Proceedings of the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), 2022.

A Paper Accepted for KDD 2022

Our paper proposing a new conditional VAE (CVAE) that acquires task-invariant latent variables across different tasks has been accepted to KDD2022.
Hiroshi Takahashi, Tomoharu Iwata, Atsutoshi Kumagai, Sekitoshi Kanai, Masanori Yamada, Yuuki Yamanaka, Hisashi Kashima.
Learning Optimal Priors for Task-Invariant Representations in Variational Autoencoders.
In Proceedings of the 28th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 2022.

A Paper Accepted for ICPR 2022


Our paper proposing a method for integrating responses collected by crowdsourcing when observation bias exists has been accepted by ICPR 2022.
Ryosuke Ueda, Koh Takeuchi, Hisashi Kashima.
Mitigating Observation Biases in Crowdsourced Label Aggregation.
In Proceedings of the 26th International Conference on Pattern Recognition (ICPR), 2022.

A Paper Accepted for UAI 2022

A paper proposing a new feature selection method for intervention effect estimation has been accepted by UAI 2022.
Yoichi Chikahara, Makoto Yamada, Hisashi Kashima.
Feature Selection for Discovering Distributional Treatment Effect Modifiers.
In Proceedings of the 38th Conference on Uncertaintly in Artificial Intelligence (UAI), 2022.

A Paper Accepted for Scientific Reports

A paper proposing a method for predicting the physical properties of compounds by correcting the experimental biases has been accepted for publication in Scientific Reports.
Yang Liu, Hisashi Kashima.
Chemical Property Prediction Under Experimental Biases.
Scientific Reports, 2022.

A paper accepted for Expert Systems with Applications (ESWA)

Our paper on learning a classifier with near-miss (weak positive) labels was accepted for the journal of Expert Systems with Applications (ESWA).
  • Akira Tanimoto, So Yamada, Takashi Takenouchi, Masashi Sugiyama, Hisashi Kashima.
    Improving Imbalanced Classification Using Near-miss Instances.
    Expert Systems with Applications (ESWA), 2022.
  • A Paper Accepted for Machine Learning Journal

    A paper proposing a method to predict spatio-temporal events with high accuracy by deepening the Hawkes process, a point process model, using CNN has been accepted for publication in Machine Learning (Special Issue of ECML PKDD).
  • Maya Okawa, Tomoharu Iwata, Yusuke Tanaka, Hiroyuki Toda, Takeshi Kurashima, Hisashi Kashima.
    Context-aware Spatio-temporal Event Prediction via Convolutional Hawkes Processes.
    Machine Learning, 2022.
  • Two Papers Accepted for AISTATS 2022

    Two papers were accepted for AISTATS 2022
    • Yuki Takezawa, Ryoma Sato, Zornitsa Kozareva, Sujith Ravi, Makoto Yamada
      Fixed Support Tree-Sliced Wasserstein Barycenter
      In Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (AISTATS 2022)
    • Benjamin Poignard, Peter Naylor, Héctor Climente, Makoto Yamada
      Feature Screening with Kernel Knockoff
      In Proceedings of the 25th International Conference on Artificial Intelligence and Statistics (AISTATS 2022)