Publications

1. I-J. Chen, M. Aapro, A. Kipnis, A. Ilin, P. Liljeroth and A. S. Foster (2022)
   Precise atom manipulation through deep reinforcement learning. [arxiv]
   Nature Communications.
   
   
2. O. Vikström and A. Ilin (2022)
   Learning explicit object-centric representations with vision transformers. [arxiv]
   NeurIPS worskhop on Vision Transformers.
   
   
3. K. Haitsiukevich and A. Ilin (2022)
   Learning trajectories of Hamiltonian systems with neural networks. [arxiv]
   International Conference on Artificial Neural Networks (ICANN 2022).
   
   
4. Y. Zhao, R. Boney, A. Ilin, J. Kannala, J. Pajarinen (2022)
   Adaptive Behavior Cloning Regularization for Stable Offline-to-Online Reinforcement Learning. [arxiv]
   European Symposium on Artificial Neural Networks (ESANN 2022).
   
   
5. S. Spilsbury and A. Ilin (2022)
   Compositional generalization in grounded language learning via induced model sparsity. [arxiv]
   NAACL Student Research Workshop 2022.
   
   
6. A. Polis and A. Ilin (2022)
   A relational model for one-shot classification of images and pen strokes. [link]
   Neurocomputing.
   
   
7. N. Oinonen, L. Kurki, A. Ilin and A. Foster (2022)
   Molecule graph reconstruction from Atomic Force Microscopy images with machine learning. [link]
   MRS Bulletin, 2022.
   
   
8. M. Saman Booy, A. Ilin and P. Orponen (2022)
   RNA secondary structure prediction with convolutional neural networks. [biorxiv]
   BMC Bioinformatics, 2022.
   
   
9. A. Viitala, R. Boney, Y. Zhao, A. Ilin, J. Kannala (2021)
   Learning to drive (L2D) as a low-cost benchmark for real-world reinforcement learning. [arxiv]
   International Conference on Advanced Robotics (ICAR 2021).
   
   
10. K. Kujanpää, W. Victor and A. Ilin (2021)
    Automating privilege escalation with deep reinforcement learning. [arxiv]
    ACM Workshop on Artificial Intelligence and Security (AISec 2021).
    
    
11. K. Haitsiukevich, S. Bergman, C. de Araujo Filho, F. Corona and A. Ilin (2021)
    A grid-structured model of tubular reactors. [arxiv]
    IEEE International Conference on Industrial Informatics (INDIN 21).
    
    
12. A. Polis and A. Ilin (2021)
    A relational model for one-shot classification. [arxiv]
    European Symposium on Artificial Neural Networks (ESANN 2021).
    
    
13. A. Keurulainen, I. Westerlund, A. Kwiatkowski, S. Kaski and A. Ilin (2021)
    Behaviour-conditioned policies for cooperative reinforcement learning tasks. [arxiv]
    International Conference on Artificial Neural Networks and Machine Learning (ICANN 2021).
    
    
14. A. Keurulainen, I. Westerlund, S. Kaski and A. Ilin (2021)
    Learning to assist agents by observing them. [arxiv]
    International Conference on Artificial Neural Networks and Machine Learning (ICANN 2021).
    
    
15. R. Boney, A. Ilin, J. Kannala and J. Seppänen (2021)
    Learning to play imperfect-information games by imitating an oracle planner. [arxiv]
    IEEE Transactions on Games.
    
    
16. Y. Kong, D. Petrov, V. Räisänen and A. Ilin (2021)
    Path-link graph neural network for IP network performance prediction. [link]
    IFIP/IEEE International Symposium on Integrated Network Management.
    
    
17. K. Palkama, L. Juvela and A. Ilin (2020)
    Conditional spoken digit generation with StyleGAN. [arxiv]
    INTERSPEECH 2020.
    
    
18. J. Tulensalo, J. Seppänen and A. Ilin (2020)
    An LSTM model for power grid loss prediction. [link]
    Electric Power Systems Research.
    
    
19. R. Boney, J. Kannala, A. Ilin (2019)
    Regularizing model-based planning with energy-based models. [arxiv]
    Conference on Robot Learning (CoRL 2019).
    
    
20. R. Boney, N. Di Palo, M. Berglund, A. Ilin, J. Kannala, A. Rasmus and H. Valpola (2019)
    Regularizing trajectory optimization with denoising autoencoders [arxiv]
    Conference on Neural Information Processing Systems (NeurIPS 2019).
    
    
21. R. Boney and A. Ilin (2019)
    Active one-shot learning with Prototypical Networks. [arxiv]
    European Symposium on Artificial Neural Networks (ESANN 2019).
    
    
22. R. Boney and A. Ilin (2017)
    Semi-supervised few-shot learning with MAML. [link]
    ICLR workshop 2017.
    
    
23. R. Boney and A. Ilin (2017)
    Semi-supervised few-shot learning with Prototypical Networks. [arxiv]
    NeurIPS workshop on meta-learning.
    
    
24. I. Prémont-Schwarz, A. Ilin, T. Hao, A. Rasmus, R. Boney and H. Valpola (2017)
    Recurrent Ladder networks [arxiv]
    Conference on Neural Information Processing Systems (NeurIPS 2017).
    
    
25. M. Abbas, A. Ilin, A. Solonen, J. Hakkarainen, E. Oja and H. Jarvinen (2016)
    Empirical evaluation of Bayesian optimization in parametric tuning of chaotic systems.
    International Journal for Uncertainty Quantification.
    
    
26. K. Cho, T. Raiko, A. Ilin and J. Karhunen (2015)
    How to pretrain deep Boltzmann machines in two stages.
    Artificial Neural Networks.
    
    
27. J. Luttinen, T. Raiko and A. Ilin.
    Linear state-space model with time-varying dynamics. [pdf]
    European Conference on Machine Learning (ECML 2014).
    
    
28. A. Solonen, J. Hakkarainen, A. Ilin, M. Abbas and A. Bibov (2014)
    Estimating model error covariance matrix parameters in extended Kalman filtering.
    Nonlinear Processes in Geophysics.
    
    
29. K. Cho, T. Raiko, A. Ilin and J. Karhunen (2013)
    A two-stage pretraining algorithm for deep Boltzmann machines. [pdf]
    International Conference on Artificial Neural Networks (ICANN 2013).
    
    
30. K. Cho, T. Raiko and A. Ilin (2013)
    Enhanced gradient for training restricted Boltzmann machines. [link]
    Neural Computation.
    
    
31. K. Cho, T. Raiko, and A. Ilin (2013)
    Gaussian-Bernoulli deep Boltzmann machine. [pdf]
    International Joint Conference on Artificial Neural Networks (IJCNN 2013).
    
    
32. J. Hakkarainen, A. Solonen, A. Ilin, J. Susiluoto, M. Laine, H. Haario and H. Järvinen (2013)
    A dilemma of the uniqueness of weather and climate model closure parameters.
    Tellus A: Dynamic Meteorology and Oceanography.
    
    
33. S. Keronen, K. Cho, T. Raiko, A. Ilin and Kalle Palomäki (2013)
    Gaussian-Bernoulli restricted Boltzmann machines and automatic feature extraction for noise robust missing data mask estimation.
    International Conference on Acoustics, Speech and Signal Processing (ICASSP 2013).
    
    
34. J. Luttinen and A. Ilin (2012)
    Efficient Gaussian process inference for short-scale spatio-temporal modeling. [pdf]
    International Conference on Artificial Intelligence and Statistics (AISTATS 2012).
    
    
35. K. Cho, A. Ilin and T. Raiko (2012)
    Tikhonov-type regularization for restricted Boltzmann machines. [pdf]
    International Conference on Artificial Neural Networks (ICANN 2012).
    
    
36. T. Hao, T. Raiko, A. Ilin and J. Karhunen (2012)
    Gated Boltzmann machine in texture modeling. [pdf]
    International Conference on Artificial Neural Networks (ICANN 2012).
    
    
37. J. Luttinen, A. Ilin and J. Karhunen (2012)
    Bayesian robust PCA of incomplete data. [link]
    Neural Processing Letters.
    
    
38. J. Hakkarainen, A. Ilin, A. Solonen, M. Laine, H. Haario, J. Tamminen, E. Oja, and H. Järvinen (2012)
    On closure parameter estimation in chaotic systems.
    Nonlinear processes in Geophysics.
    
    
39. K. Cho, T. Raiko and A. Ilin (2011)
    Enhanced gradient and adaptive learning rate for training restricted Boltzmann machines. [pdf]
    International Conference on Machine Learning (ICML 2011)
    
    
40. K. Cho, A. Ilin and T. Raiko (2011)
    Improved learning of Gaussian-Bernoulli restricted Boltzmann machines. [pdf]
    International Conference on Artificial Neural Networks (ICANN 2011)
    
    
41. J. Hegedüs, Y. Miche, A. Ilin and A. Lendasse (2011)
    Methodology for behavioral-based malware analysis and detection using random projections and k-nearest neighbors classifiers.
    International Conference on Computational Intelligence and Security.
    
    
42. K. Cho, T. Raiko and A. Ilin (2010)
    Parallel tempering is efficient for learning restricted Boltzmann machines. [pdf]
    International Joint Conference on Artificial Neural Networks (IJCNN 2010).
    
    
43. J. Luttinen and A. Ilin (2010)
    Transformations in variational Bayesian factor analysis to speed up learning.
    Neurocomputing.
    
    
44. A. Ilin and T. Raiko (2010)
    Practical approaches to principal component analysis in the presence of missing values. [link]
    Journal of Machine Learning Research.
    
    
45. J. Luttinen and A. Ilin (2009)
    Variational Gaussian-Process factor analysis for modeling spatio-temporal data. [pdf]
    Conference on Neural Information Processing Systems (NeurIPS 2009).
    
    
46. L. Kozma, A. Ilin and T. Raiko (2009)
    Binary principal component analysis in the Netflix collaborative filtering task.
    IEEE International Workshop on Machine Learning for Signal Processing (MLSP 2009).
    
    
47. A. Ilin and A. Kaplan (2009)
    Bayesian PCA for reconstruction of historical seasurface temperatures.
    International Joint Conference on Neural Networks.
    
    
48. J. Luttinen, A. Ilin and T. Raiko (2009)
    Transformations for variational factor analysis to speed up learning.
    European Symposium on Artificial Neural Networks (ESANN 2009)
    
    
49. J. Luttinen, A. Ilin and J. Karhunen (2009)
    Bayesian robust PCA for incomplete data.
    International Conference on Independent Component Analysis and Signal Separation.
    
    
50. T. Raiko, A. Ilin and J. Karhunen (2007)
    Principal component analysis for large scale problems with lots of missing values.
    European Conference on Machine Learning (ECML 2007).
    
    
51. A. Honkela, H. Valpola, A. Ilin and J. Karhunen (2007)
    Blind separation of nonlinear mixtures by variational Bayesian learning. [pdf]
    Digital Signal Processing.
    
    
52. T. Raiko, A. Ilin and J. Karhunen (2007)
    Principal component analysis for sparse high-dimensional data.
    International Conference on Neural Information Processing (ICONIP 2007).
    
    
53. A. Ilin, H. Valpola and E. Oja (2006)
    Exploratory analysis of climate data using source separation methods.
    Neural Networks.
    
    
54. A. Ilin, H. Valpola and E. Oja (2006)
    Extraction of components with structured variance.
    International Joint Conference on Neural Network (IJCNN 2006).
    
    
55. S. Borisov, A. Ilin, R. Vigário and E. Oja (2006)
    Comparison of BSS methods for the detection of α-activity components in EEG.
    International Conference on Independent Component Analysis and Signal Separation.
    
    
56. Alexander Ilin (2006). Independent dynamics subspace analysis.
    European Symposium on Artificial Neural Networks (ESANN 2006).
    
    
57. A. Ilin and H. Valpola (2005)
    On the effect of the form of the posterior approximation in variational learning of ICA models [link]
    Neural Processing Letters.
    
    
58. A. Ilin and H. Valpola (2005)
    Frequency-based separation of climate signals.
    European Conference on Principles of Data Mining and Knowledge Discovery (PKDD 2005).
    
    
59. A. Ilin, H. Valpola and Erkki Oja (2005)
    Semiblind source separation of climate data detects El Niño as the component with the highest interannual variability.
    International Joint Conference on Neural Networks (IJCNN 2005).
    
    
60. A. Ilin, H. Valpola and E. Oja (2004)
    Nonlinear dynamical factor analysis for state change detection. [link]
    IEEE Transactions on Neural Networks.
    
    
61. A. Ilin and A. Honkela (2004)
    Post-nonlinear independent component analysis by variational Bayesian learning.
    International Conference on Independent Component Analysis and Signal Separation.
    
    
62. A. Ilin, S. Achard and C. Jutten (2004)
    Bayesian versus constrained structure approaches for source separation in post-nonlinear mixtures.
    International Joint Conference on Neural Networks (IJCNN 2004).
    
    
63. H. Valpola, E. Oja, A. Ilin, A. Honkela and J. Karhunen (2003)
    Nonlinear blind source separation by variational Bayesian learning.
    IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences.
    
    
64. A. Ilin and H. Valpola (2003)
    On the effect of the form of the posterior approximation in variational learning of ICA models.
    International Conference on Independent Component Analysis and Blind Signal Separation.
    
    
65. A. Iline, H. Valpola and E. Oja (2001)
    Detecting process state changes by nonlinear blind source separation.
    International Conference on Independent Component Analysis and Blind Signal Separation.