Learn from the brain, but not be constrained by it. Artificial intelligence research has been profoundly influenced by neuroscience, inspiring advancements in areas such as the representation of visual stimuli [1, 2] and visual attention mechanisms [3, 4]. However, as one of the ultimate goals of AI is to surpass human capabilities [5], modern algorithms do not need to strictly mimic biological processes [6]. Our research philosophy embraces this balance: leveraging biological insights to inform computational algorithms where appropriate, while maintaining the freedom to explore beyond biological constraints.

Latest Research

A recent example of our bio-inspired machine learning work focuses on the Sequential Probability Ratio Test (SPRT) [7]. Originally proposed by Abraham Wald, SPRT was later found to describe the activity of neurons in the primate parietal lobe [8], which are believed to play a role in evidence accumulation during complex decision-making tasks. Building on this foundation, we expanded Wald’s algorithm to address more complex scenarios by integrating deep neural networks with advanced density ratio estimation techniques [9]. This extension, SPRT-TANDEM, demonstrates competitive speed and accuracy across various real-world applications.

For those interested, we invite you to explore our introductory article on GitHub and delve deeper into the original research paper [10].

Research Interests

• Bio-inspired machine learning
• Sequential Probability Ratio Test (SPRT)
• Density ratio estimation
• Face
  • Face recognition (computer vision)
  • Face recognition (visual neuroscience)
  • Face presentation attack detection (PAD)
    
    

News

[Jan. 23rd, 2025]　　

• Our paper, “Learning the Optimal Stopping for Early Classification within Finite Horizons via Sequential Probability Ratio Test” was accepted to ICLR 2025.　　

[Apr. 1st, 2024]　　

• Akinori F. Ebihara has been promoted to Principal Researcher.

[Feb. 16th, 2023]　　

• Our paper, “Toward Asymptotic Optimality: Sequential Unsupervised Regression of Density Ratio for Early Classification” was accepted to ICASSP 2023.

[Apr. 27th, 2022]　　

• Our paper, “Convolutional Neural Networks for Time-dependent Classification of Variable-length Time Series” was accepted to WCCI 2021 (IJCNN track) as an oral presentation.　　

[Jun. 2nd, 2021]　　

• Our paper, “Joint Feature Distribution Alignment Learning for NIR-VIS and VIS-VIS Face Recognition” was accepted to IJCB 2021.　　

[May 8th, 2021]　　

• Our paper, “Sequential Density Ratio Matrix Estimation for Speed-Accuracy Optimization” was accepted to ICML 2021.　　

[May 7th, 2021]

• Nikkei (日本経済新聞) and Nikkei XTech featured articles about our algorithm, SPRT-TANDEM.
• NEC released an official press release about the SPRT-TANDEM.

[Apr. 1st, 2021]

• Akinori F. Ebihara has been promoted to “Special Researcher.”
  *A few distinguished researchers that are selected to enter a performance-based reward system instead of the traditional seniority wage system.

[Jan. 13th, 2021]

• Our paper, “Sequential Density Ratio Estimation for Simultaneous Optimization of Speed and Accuracy” was accepted as a Spotlight presentation (top 5.57%) at ICLR 2021.

[Oct. 1st, 2020]

• Our paper, “Specular- and Diffuse-reflection-based Face Spoofing Detection for Mobile Devices” won IJCB 2020 Google Best Paper Award.

References

1. Hubel, D. H., Wiesel, T. N. RECEPTIVE FIELDS AND FUNCTIONAL ARCHITECTURE IN TWO NONSTRIATE VISUAL AREAS (18 AND 19) OF THE CAT. Journal of Neurophysiology 1965, 28 (2), 229–289.
2. Fukushima, K. Neocognitron: A Self-Organizing Neural Network Model for a Mechanism of Pattern Recognition Unaffected by Shift in Position. Biol. Cybernetics 1980, 36 (4), 193–202.
3. Posner, M. I., Petersen, S. E. The Attention System of the Human Brain. Annu. Rev. Neurosci. 1990, 13 (1), 25–42.
4. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., Polosukhin, I. Attention Is All You Need. 11.
5. Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A., Hubert, T., Baker, L., Lai, M., Bolton, A., Chen, Y., Lillicrap, T., Hui, F., Sifre, L., van den Driessche, G., Graepel, T., Hassabis, D. Mastering the Game of Go without Human Knowledge. Nature 2017, 550 (7676), 354–359.
6. Nonaka, S., Majima, K., Aoki, S. C., Kamitani, Y. Brain Hierarchy Score: Which Deep Neural Networks Are Hierarchically Brain-Like? bioRxiv 2020.
7. Wald, A. Sequential Analysis. Wiley, Hoboken, NJ, USA, 1947.
8. Kira, S., Yang, T., Shadlen, M. N. A Neural Implementation of Wald’s Sequential Probability Ratio Test. Neuron 2015, 85 (4), 861–873.
9. Sugiyama, M., Suzuki, T., Kanamori, T. Density ratio estimation in machine learning. Cambridge University Press, 2012.
10. Ebihara, A.F., Miyagawa, T., Sakurai, K., Imaoka, H. Sequential Density Ratio Estimation for Simultaneous Optimization of Speed and Accuracy. ICLR 2021.