Takanori MAEHARA

• Senior Software Engineer, Roku, Inc.
• 2 Cambridge Science Park Milton Rd, Milton, Cambridge CB4 0AF
• tmaehara@acm.org (personal)

Takanori Maehara is a Machine Learning Engineer specializing in Search and Recommendation systems. He is also a Researcher in Theoretical Computer Science focusing on Discrete Mathematics and Theoretical Machine Learning. After receiving his PhD from the University of Tokyo in 2012, he studied discrete optimisation and learning theory in academia. He worked as a Post-Doctoral Researcher at National Institute of Informatics, an Associate Professor at Shizuoka University, and a Unit Leader at RIKEN Center for Advanced Intelligence Project. In 2020, he transitioned to tech industry to gain practical machine learning experience. He worked as a Software Engineer at Facebook (now known as Meta) and is now working as a Senior Software Engineer at Roku.

Brief CV

3-Page CV (pdf), Full CV (pdf)

Publications (all peer-reviewed)

1. Takanori Maehara, Hoang NT (2024): “Deep homomorphism network”. Proceedings of the 38th Annual Conference on Neural Information Processing Systems (NeurIPS'24), Vancouver, Canada, December 10–15, 2024.[bibtex] 
2. Takanori Maehara, So Nakashima (2022): “Rank axiom of modular supermatroids: A connection with directional DR submodular functions”. Advances in Applied Mathematics, vol. 134, pp. 102304.[bibtex] 
3. Takanori Maehara, Hoang NT (2021): “Learning on Random Balls is Sufficient for Estimating (Some) Graph Parameters”. Proceedings of the 35th Annual Conference on Neural Information Processing Systems (NeurIPS'21), Online, December 7–10, 2021.[bibtex] 
4. Soh Kumabe, Takanori Maehara (2021): “Prophet Secretary for k-Knapsack and l-Matroid Intersection via Continuous Exchange Property”. Proceedings of the 32nd International Workshop on Combinatorial Algorithms (IWOCA'21), Online, July 5–7, 2021.[bibtex] 
5. Takanori Maehara, So Nakashima, Yutaro Yamaguchi (2021): “Multiple Knapsack-Constrained Monotone DR-Submodular Maximization on Distributive Lattice”. Mathematical Programming.[bibtex] 
6. Hoang NT, Takanori Maehara, Tsuyoshi Murata (2021): “Revisiting Graph Neural Networks: Graph Filtering Perspective”. Proceedings of the 25th International Conference on Pattern Recognition (ICPR'20), Online, January 10–15, 2021.[bibtex] 
7. Mario Coutino, Sundeep Prabhakar Chepuri, Takanori Maehara, Geert Leus (2020): “Fast Spectral Approximation of Structured Graphs with Applications to Graph Filtering”. Algorithms, vol. 13, no. 9, pp. 214.[bibtex] 
8. Yoshifumi Seki, Takanori Maehara (2020): “A Method to Anonymize Business Metrics to Publishing Implicit Feedback Datasets”. Proceedings of the 14th ACM Conference on Recommender Systems (RecSys'20), pp. 4–12.[bibtex] 
9. Hoang NT, Takanori Maehara (2020): “Graph Homomorphism Convolution”. Proceedings of the 37th International Conference on Machine Learning (ICML'20), pp. 10552–10562.[bibtex] 
10. Mario Coutino, Elvin Isufi, Takanori Maehara, Geert Leus (2020): “State-Space Based Network Topology Identification”. Proceedings of the 28th European Signal Processing Conference (EUSIPCO'20), Online, January 18–22, 2021, pp. 1055-1059.[bibtex] 
11. Mario Coutino, Elvin Isufi, Takanori Maehara, Geert Leus (2020): “State-Space Network Topology Identification From Partial Observations”. IEEE Transactions on Signal and Information Processing over Networks, pp. 211–225.[bibtex] 
12. Soh Kumabe, Takanori Maehara (2020): “Convexity of b-Matching Game”. Proceedings of the 29th International Joint Conference on Artificial Intelligence and the 17th Pacific Rim International Conference on Artificial Intelligence (IJCAI-PRICAI'20), Online, January 7–15, 2021, pp. 261–267.[bibtex] 
13. Yoichi Sasaki, Takanori Maehara, Takumi Akazaki, Kazeto Yamamoto, Kunihiko Sadamasa (2020): “Solving Weighted Abduction via Max-SAT Solvers”. Proceedings of the 33rd International FLAIRS Conference (FLAIRS'20), pp. 142–147.[bibtex] 
14. Soh Kumabe, Takanori Maehara (2020): “Convexity of Hypergraph Matching Game”. Proceedings of the 19th International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS'20), Online, May 9–13, 2020, pp. 663–671.[bibtex] 
15. Kazuto Fukuchi, Satoshi Hara, Takanori Maehara (2020): “Faking Fairness via Stealthily Biased Sampling”. Proceedings of the 34th AAAI Conference on Artificial Intelligence (AAAI'20), Special Track on AI for Social Impact, New York, New York, USA, February 7–12, 2020, pp. 412–419.[bibtex] 
16. Satoshi Hara, Atsushi Nitanda, Takanori Maehara (2019): “Data Cleansing for Models Trained with SGD”. Proceedings of the 33rd Annual Conference on Neural Information Processing Systems (NeurIPS'19), Vancouver, Canada, December 8–14, 2019, pp. 4215–4224.[bibtex] 
17. Satoshi Hara, Takanori Maehara (2019): “Convex Hull Approximation of Nearly Optimal Lasso Solutions”. Proceedings of the 16th Pacific Rim International Conference on Artificial Intelligence (PRICAI'19), anuca Island, Cuvu, Fiji, August 26–30, 2019, pp. 350–363.[bibtex] 
18. Junjie Chen, Takanori Maehara (2019): “Chance-Constrained Submodular Knapsack Problem”. Proceedings of the 25th International Computing and Combinatorics Conference (COCOON'19), Xian, China, July 29–31, 2019, pp. 103–114.[bibtex] 
19. Masakazu Ishihata, Takanori Maehara (2019): “Exact Bernoulli Scan Statistics using Binary Decision Diagrams”. Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI'19), Macau, China, August 10–16, 2019, pp. 5737–5743.[bibtex] 
20. Fukunaga, Takuro, Maehara, Takanori (2019): “Computing a tree having a small vertex cover”. Theoretical Computer Science, vol. 791, pp. 48–61.[bibtex] 
21. Takanori Maehara, Yutaro Yamaguchi (2019): “Stochastic packing integer programs with few queries”. Mathematical Programming, Series A, pp. 1–34.[bibtex] 
22. Mohammed Alsuhaibani, Takanori Maehara, Danushka Bollegala (2019): “Joint Learning of Hierarchical Word Embeddings from a Corpus and a Taxonomy”. Proceedings of the 1st Conference on Automated Knowledge Base Construction (AKBC'19), University of Massachusetts Amherst, United States, May 20–22, 2019.[bibtex] 
23. Ben Chugg, Takanori Maehara (2019): “Submodular Stochastic Probing with Prices”. Proceedings of the 6th International Conference on Control, Decision and Information Technologies (CoDIT'19), Paris, France, April 23–25, 2019, pp. 60–66.[bibtex] 
24. Soh Kumabe, Takanori Maehara, Ryoma Sinya (2019): “Linear Pseudo-Polynomial Factor Algorithm for Automaton Constrained Tree Knapsack Problem”. Proceedings of the 13th International Conference and Workshops on Algorithms and Computation (WALCOM'19), Guwahati, India, February 27–March 2, 2019, pp. 248–260.[bibtex] 
25. So Nakashima, Takanori Maehara (2019): “Subspace Selection via DR-Submodular Maximization on Lattices”. Proceedings of the 33rd AAAI Conference on Artificial Intelligence (AAAI'19), Honolulu, Hawaii, January 27–February 1, 2019, pp. 4618–4625.[bibtex] 
26. Takanori Maehara, Yuma Inoue (2019): “Group Decision Diagram (GDD): A Compact Representation for Permutations”. Proceedings of the 33rd AAAI Conference on Artificial Intelligence (AAAI'19), Honolulu, Hawaii, United States, January 27–February 1, 2019, pp. 2986–2994.[bibtex] 
27. Taro Takaguchi, Takanori Maehara, Ken-ichi Kawarabayashi, Masashi Toyoda (2018): “Existence of outsiders as a characteristic of online communication networks”. Network Science, vol. 6, no. 4, pp. 431–447.[bibtex] 
28. Mario Coutino, Elvin Isufi, Takanori Maehara, Geert Leus (2018): “On The Limits of Finite-Time Distributed Consensus Through Successive Local Linear Operations”. Proceedings of the 52nd Asilomar Conference on Signals, Systems, and Computers (ACSSC'18), Pacific Grove, CA, USA, October 28–31, 2018, pp. 993–997.[bibtex] 
29. Danushka Bollegala, Vincent Atanasov, Takanori Maehara, Ken-ichi Kawarabayashi (2018): “ClassiNet - Predicting Missing Features for Short-Text Classification”. ACM Transactions on Knowledge Discovery from Data, vol. 12, no. 5, pp. 55:1–55:29.[bibtex] 
30. Tatsunori Taniai, Takanori Maehara (2018): “Neural Inverse Rendering for General Reflectance Photometric Stereo”. Proceedings of the 35th International Conference on Machine Learning (ICML'18), Stockholm, Sweden, July 10-15, 2018, pp. 4864–4873.[bibtex] 
31. Takanori Maehara, Atsuhiro Narita, Jun Baba, Takayuki Kawabata (2018): “Optimal Bidding Strategy for Brand Advertising”. Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI'18), Stockholm, Sweden, July 13–19, 2018, pp. 424–432.[bibtex] 
32. Takanori Maehara, Naoki Marumo, Kazuo Murota (2018): “Continuous relaxation for discrete DC programming”. Mathematical Programming Series B, vol. 169, no. 1, pp. 199–219.[bibtex] 
33. Masaaki Imaizumi, Takanori Maehara, Yuichi Yoshida (2018): “Statistically Efficient Estimation for Non-Smooth Probability Densities”. International Conference on Artificial Intelligence and Statistics (AISTATS'18), Playa Blanca, Lanzarote, Canary Islands, Spain, April 9–11, 2018, pp. 978–987.[bibtex] 
34. Alsuhaibani, Mohammed, Bollegala, Danushka, Maehara, Takanori, Kawarabayashi, Ken-ichi (2018): “Jointly learning word embeddings using a corpus and a knowledge base”. PloS ONE, vol. 13, no. 3, pp. e0193094.[bibtex] 
35. Satoshi Takabe, Takanori Maehara, Koji Hukushima (2018): “Typical approximation performance for maximum coverage problem”. Physical Review E, vol. 97, no. 2, pp. 022138.[bibtex] 
36. Takayuki Osogami, Rudy Raymond, Akshay Goel, Tomoyuki Shirai, Takanori Maehara (2018): “Dynamic Determinantal Point Processes”. Proceedings of the 32nd AAAI Conference on Artificial Intelligence (AAAI'18), New Orleans, Louisiana, USA, February 2–7, 2018, pp. 3868–3875.[bibtex] 
37. Takanori Maehara, Yutaro Yamaguchi (2018): “Stochastic Packing Integer Programs with Few Queries”. Proceedings of the 29th Annual ACM-SIAM Symposium on Discrete Algorithms (SODA'18), New Orleans, LA, USA, January 7–10, 2018, pp. 293–310.[bibtex] 
38. Takanori Maehara, Yasushi Kawase, Hanna Sumita, Katsuya Tono, Ken-ichi Kawarabayashi (2017): “Optimal Pricing for Submodular Valuations with Bounded Curvature”. Proceedings of the 31st AAAI Conference on Artificial Intelligence (AAAI'17), San Francisco, California, USA, February 4–9, 2017, pp. 622–628.[bibtex] 
39. Satoshi Hara, Takanori Maehara (2017): “Enumerate Lasso Solutions for Feature Selection”. Proceedings of the 31st AAAI Conference on Artificial Intelligence (AAAI'17), San Francisco, California, USA, February 4–9, 2017, pp. 1985–1991.[bibtex] 
40. Daisuke Hatano, Takuro Fukunaga, Takanori Maehara, Ken-ichi Kawarabayashi (2017): “Scalable Algorithm for Higher-Order Co-Clustering via Random Sampling”. Proceedings of the 31st AAAI Conference on Artificial Intelligence (AAAI'17), San Francisco, California, USA, February 4–9, 2017, pp. 1992–1999.[bibtex] 
41. Masaaki Imaizumi, Takanori Maehara, Kohei Hayashi (2017): “On Tensor Train Rank Minimization : Statistical Efficiency and Scalable Algorithm”. Proceedings of the 31st Annual Conference on Neural Information Processing Systems (NIPS'17), Long Beach, California, USA, December 4–9 2017, pp. 3933–3942.[bibtex] 
42. Takanori Maehara, Hirofumi Suzuki, Masakazu Ishihata (2017): “Exact Computation of Influence Spread by Binary Decision Diagrams”. Proceedings of the 26th International Conference on World Wide Web (WWW'17), Perth, Australia, April 3–7, 2017, pp. 947–956.[bibtex] 
43. Ryosuke Nishi, Taro Takaguchi, Keigo Oka, Takanori Maehara, Masashi Toyoda, Ken-ichi Kawarabayashi, Naoki Masuda (2016): “Reply trees in Twitter: data analysis and branching process models”. Social Network Analysis and Mining, vol. 6, no. 1, pp. 26:1–26:13.[bibtex] 
44. Takuro Fukunaga, Takanori Maehara (2016): “Computing a Tree Having a Small Vertex Cover”. Proceedings of the 10th International Conference on Combinatorial Optimization and Applications (COCOA'16), Hong Kong, China, December 16–18, 2016, pp. 77–91.[bibtex] 
45. Takanori Maehara, Kohei Hayashi, Ken-ichi Kawarabayashi (2016): “Expected Tensor Decomposition with Stochastic Gradient Descent”. Proceedings of the 30th AAAI Conference on Artificial Intelligence (AAAI'16), Phoenix, Arizona, USA, February 12–17, 2016, pp. 1919–1925.[bibtex] 
46. Danushka Bollegala, Mohammed Alsuhaibani, Takanori Maehara, Ken-ichi Kawarabayashi (2016): “Joint Word Representation Learning Using a Corpus and a Semantic Lexicon”. Proceedings of the 30th AAAI Conference on Artificial Intelligence (AAAI'16), Phoenix, Arizona, USA., February 12–17, 2016, pp. 2690–2696.[bibtex] 
47. Kohei Hayashi, Takanori Maehara, Masashi Toyoda, Ken-ichi Kawarabayashi (2015): “Real-Time Top-R Topic Detection on Twitter with Topic Hijack Filtering”. Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD'15), Sydney, New South Wales, Australia, August 10–13, 2015, pp. 417–426.[bibtex] 
48. Naoto Ohsaka, Takanori Maehara, Ken-ichi Kawarabayashi (2015): “Efficient PageRank Tracking in Evolving Networks”. Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD'15), Sydney, New South Wales, Australia, August 10–13, 2015, pp. 875–884.[bibtex] 
49. Danushka Bollegala, Takanori Maehara, Ken-ichi Kawarabayashi (2015): “Unsupervised Cross-Domain Word Representation Learning”. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics (ACL'15), July 26–31, 2015, Beijing, China, pp. 730–740.[bibtex] 
50. Takanori Maehara, Akihiro Yabe, Ken-ichi Kawarabayashi (2015): “Budget Allocation Problem with Multiple Advertisers: A Game Theoretic View”. Proceedings of the 32nd International Conference on Machine Learning (ICML'15), Lille, France, July 6–11, 2015, pp. 428–437.[bibtex] 
51. Danushka Bollegala, Takanori Maehara, Ken-ichi Kawarabayashi (2015): “Embedding Semantic Relations into Word Representations”. Proceedings of the 24th International Joint Conference on Artificial Intelligence (IJCAI'15), Buenos Aires, Argentina, July 25–31, 2015, pp. 1222–1228.[bibtex] 
52. Takanori Maehara, Kazuo Murota (2015): “Valuated matroid-based algorithm for submodular welfare problem”. Annals of Operations Research, vol. 229, no. 1, pp. 565–590.[bibtex] 
53. Takanori Maehara, Kazuo Murota (2015): “A framework of discrete DC programming by discrete convex analysis”. Mathematical Programming, Series A, vol. 152, no. 1-2, pp. 435–466.[bibtex] 
54. Takanori Maehara (2015): “Risk averse submodular utility maximization”. Operations Research Letters, vol. 43, no. 5, pp. 526–529.[bibtex] 
55. Takanori Maehara, Naoki Marumo, Kazuo Murota (2015): “Continuous Relaxation for Discrete DC Programming”. Proceedings of the 3rd International Conference on Modelling, Computation and Optimization in Information Systems and Management Sciences (MCO'15), Metz, France, May 11–13, 2015, pp. 181–190.[bibtex] 
56. Yasushi Kawase, Takanori Maehara, Ken-ichi Kawarabayashi (2015): “Scalable sensor localization via ball-decomposition algorithm”. Proceedings of the 14th IFIP Networking Conference (Networking'15), Toulouse, France, May 20–22, 2015, pp. 1–9.[bibtex] 
57. Takanori Maehara, Mitsuru Kusumoto, Ken-ichi Kawarabayashi (2015): “Scalable SimRank join algorithm”. Proceedings of the 31st IEEE International Conference on Data Engineering (ICDE'15), Seoul, South Korea, April 13–17, 2015, pp. 603–614.[bibtex] 
58. Daisuke Hatano, Takuro Fukunaga, Takanori Maehara, Ken-ichi Kawarabayashi (2015): “Lagrangian Decomposition Algorithm for Allocating Marketing Channels”. Proceedings of the 29th AAAI Conference on Artificial Intelligence (AAAI'15), Austin, Texas, USA, January 25–30, 2015, pp. 1144–1150.[bibtex] 
59. Danushka Bollegala, Takanori Maehara, Yuichi Yoshida, Ken-ichi Kawarabayashi (2015): “Learning Word Representations from Relational Graphs”. Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence (AAAI'15), Austin, Texas, USA, January 25–30, 2015, pp. 2146–2152.[bibtex] 
60. Takanori Maehara, Takuya Akiba, Yoichi Iwata, Ken-ichi Kawarabayashi (2014): “Computing Personalized PageRank Quickly by Exploiting Graph Structures”. Proceedings of the VLDB Endowment (The 40th International Conference on Very Large Data Bases (VLDB'14), Hangzhou, China, September 1–5, 2014), vol. 7, no. 12, pp. 1023–1034.[bibtex] 
61. Mitsuru Kusumoto, Takanori Maehara, Ken-ichi Kawarabayashi (2014): “Scalable similarity search for SimRank”. Proceedings of the 2014 ACM SIGMOD International Conference on Management of Data (SIGMOD'14), Snowbird, Utah, USA, June 22–27, 2014, pp. 325–336.[bibtex] 
62. Takanori Maehara, Kazuo Murota (2011): “Algorithm for Error-Controlled Simultaneous Block-Diagonalization of Matrices”. SIAM Journal on Matrix Analysis Applications, vol. 32, no. 2, pp. 605–620.[bibtex] 
63. Takanori Maehara, Kazuo Murota (2020): “A numerical algorithm for block-diagonal decomposition of matrix *-algebras with general irreducible components”. Japan Journal of Industrial and Applied Mathematics, vol. 27, pp. 263–293.[bibtex] 
64. Harold W. Gutch, Takanori Maehara, Fabian J. Theis (2010): “Second Order Subspace Analysis and Simple Decompositions”. Proceedings of the 9th International Conference on Latent Variable Analysis and Signal Separation (LVA/ICA'10), St. Malo, France, September 27–30, 2010, pp. 370–377.[bibtex] 
65. Toshio Funada, Daniel D Joseph, Takanori Maehara, Susumu Yamashita (2005): “Ellipsoidal model of the rise of a Taylor bubble in a round tube”. International Journal of Multiphase Flow, vol. 31, no. 4, pp. 473–491.[bibtex] 

Tools, Softwares and Algorithm Implementations

• Implementations of Algorithms and Data Structures in Rust
• Bot that fetches RSS feeds and reposts to Bluesky. Implemented in Rust.

• M#-Convex Funtion Partition Minimization (DCP Project)
• Error Controlled Simultaneous Block-Diagonalization