Meeting #2435
Updated by Evgeniy Pavlovskiy almost 6 years ago
h1. 1 Schedule [[Seminars_schedule]] h1. 2 Requirements 25 minutes for one presentation. Main requirements to presentation: * to be prepared in LaTeX (or Jupyter Notebook with LaTeX inline), * to be short, understandable, clear and convinient, * no more than 20 minutes for content deliver and 5 for questions, * references on the last slide h1. 3 Topics Each student has to present a research and part of his thesis. Opened list of cutting-edge topics: |_.Topic|_.Link|_.Reporter|_.Scheduled| |\4=.*General*| |[ ]Zero-One shot Learning|Xian, Y., Lampert, C. H., Schiele, B., & Akata, Z. (2018). Zero-shot learning-a comprehensive evaluation of the good, the bad and the ugly. IEEE transactions on pattern analysis and machine intelligence. URL: https://ieeexplore.ieee.org/abstract/document/8413121 |*Vladislav Panferov*|| |[x]MXNet DL framework|Chen T. et al. Mxnet: A flexible and efficient machine learning library for heterogeneous distributed systems //arXiv preprint arXiv:1512.01274. – 2015. URL: https://arxiv.org/pdf/1512.01274||| |[ ] |“Why Should I Trust You?” Explaining the Predictions of Any Classifier, https://arxiv.org/pdf/1602.04938.pdf, https://github.com/marcotcr/lime|*Rohan Kumar Rathore*|?| |[ ]Manifold MixUp| Manifold Mixup: Better Representations by Interpolating Hidden States. URL: https://arxiv.org/pdf/1806.05236v4 || | |[ ]UMAP | McInnes, Leland and John Healy (2018). “UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction”. In: ArXiv e-prints. arXiv: "1802.03426 [stat.ML]":http://arxiv.org/abs/1802.03426||| |\4=.*Fluid, Oil, Physics, Chemistry*| |[ ]Data-driven predictive using field inversion|Parish, Eric J., and Karthik Duraisamy. "A paradigm for data-driven predictive modeling using field inversion and machine learning." Journal of Computational Physics 305 (2016): 758-774.|*Omid Razizadeh*|*31-Oct*| |[ ]Predicting Oil Movement in a development System Using Deep Latent Dynamic Models|URL: Video: https://www.youtube.com/watch?v=N3iV-F4aqLA? Slieds: https://bayesgroup.github.io/bmml_sem/2018/Temirchev_Metamodelling.pdf||| |\4=_.Faces| |[ ]SphereFace| SphereFace: Deep Hypersphere Embedding for Face Recognition URL: https://arxiv.org/pdf/1704.08063.pdf|| | |[x]Triplet Loss|https://arxiv.org/pdf/1503.03832.pdf||| |[x]Style transfer SotA (state-of-the-art)| A Style-Based Generator Architecture for Generative Adversarial Networks. URL: https://arxiv.org/abs/1812.04948 ||| |Performance of Word Embeddings|review and experience||| |[x]CosFace|Wang H. et al. Cosface: Large margin cosine loss for deep face recognition //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. – 2018. – С. 5265-5274. URL: https://arxiv.org/pdf/1801.09414.pdf||| |[x]Text clustering | Xu J. et al. Self-taught convolutional neural networks for short text clustering //Neural Networks. – 2017. – Т. 88. – С. 22-31. URL: https://arxiv.org/abs/1701.00185 ||| |\4=.*Quantum*| |[ ]DisCoCat model| Grefenstette E. Category-theoretic quantitative compositional distributional models of natural language semantics //arXiv preprint arXiv:1311.1539. – 2013. URL: https://arxiv.org/abs/1311.1539||| |[x]DisCoCat toy model|Gogioso S. A Corpus-based Toy Model for DisCoCat //arXiv preprint arXiv:1605.04013. – 2016. URL: https://arxiv.org/pdf/1605.04013.pdf||| |[ ]A Quantum-Theoretic Approach to Distributional Semantics|Blacoe W., Kashefi E., Lapata M. A quantum-theoretic approach to distributional semantics //Proceedings of the 2013 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. – 2013. – С. 847-857. URL: http://www.aclweb.org/anthology/N13-1105||| |Solving the Quantum Many-Body problem with ANN| Carleo G., Troyer M. Solving the quantum many-body problem with artificial neural networks //Science. – 2017. – Т. 355. – №. 6325. – С. 602-606. URL: https://arxiv.org/pdf/1606.02318||| |\4=.*Economics*| |Understanding consumer behavior|Lang T., Rettenmeier M. Understanding consumer behavior with recurrent neural networks //Workshop on Machine Learning Methods for Recommender Systems. – 2017. URL: https://doogkong.github.io/2017/papers/paper2.pdf|*Abhishek Saxena*|| https://doogkong.github.io/2017/papers/paper2.pdf||| |\4=.*Speech*| |[ ]Tacotron 2|Shen J. et al. Natural tts synthesis by conditioning wavenet on mel spectrogram predictions //2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). – IEEE, 2018. – С. 4779-4783. URL: https://arxiv.org/pdf/1712.05884.pdf||| |BERT (Google)|Devlin J. et al. Bert: Pre-training of deep bidirectional transformers for language understanding //arXiv preprint arXiv:1810.04805. – 2018. URL: https://arxiv.org/abs/1810.04805|*Nikita Nikolaev*|| https://arxiv.org/abs/1810.04805||| |\4=.*Natural Language Processing*| |[x]Universal Sentence Encoder|Cer D. et al. Universal sentence encoder //arXiv preprint arXiv:1803.11175. – 2018. URL: https://arxiv.org/pdf/1803.11175.pdf||| |[x]ULMFiT|Howard J., Ruder S. Universal language model fine-tuning for text classification //arXiv preprint arXiv:1801.06146. – 2018. URL: https://arxiv.org/pdf/1801.06146.pdf||| |ELMo|Peters M. E. et al. Deep contextualized word representations //arXiv preprint arXiv:1802.05365. – 2018. URL: http://www.aclweb.org/anthology/N18-1202|*Sergey Garmaev*|| http://www.aclweb.org/anthology/N18-1202||| |[ ]Skip-thoughts, Infersent, RandSent - Facebook|1. Kiros R. et al. Skip-thought vectors //Advances in neural information processing systems. – 2015. – С. 3294-3302. URL: https://arxiv.org/pdf/1506.06726.pdf 2. Conneau A. et al. Supervised learning of universal sentence representations from natural language inference data //arXiv preprint arXiv:1705.02364. – 2017. URL: https://arxiv.org/abs/1705.02364 3. Wieting J., Kiela D. No Training Required: Exploring Random Encoders for Sentence Classification //arXiv preprint arXiv:1901.10444. – 2019. URL: https://arxiv.org/pdf/1901.10444.pdf||| |[ ]BigARTM|Vorontsov K. et al. Bigartm: Open source library for regularized multimodal topic modeling of large collections //International Conference on Analysis of Images, Social Networks and Texts. – Springer, Cham, 2015. – С. 370-381. URL: http://www.machinelearning.ru/wiki/images/e/ea/Voron15aist.pdf||| |[x]Reinforcement Learning|Human-level control through deep reinforcement learning||| ||Selsam, D., Lamm, M., Bünz, B., Liang, P., de Moura, L., & Dill, D. L. (2018). Learning a SAT solver from single-bit supervision. arXiv preprint arXiv:1802.03685.||| h1. 4 Topics of master thesis Opened list of reports on master thesis (statement of work, review, and results): |_.Reporter|_.Topic|_.Scheduled| |\3=.1st year students| |1 to-be-listed | |-| |\3=.2st year students| |1 Razizadeh Omid | || |2 Siyoto Owen | || |3 Munyaradzi Njera | || |4 Averyanov Evgeniy | || |5 Kozinets Roman | || |6 Tagirova Elizaveta | || |7 Tsvaki Jetina | || |8 Ravi Kumar | || h1. 5 At fault These students still didn't selected a paper to report or doesn't assigned to a time slot: 1st year students * *noname*: refer, master 2nd year students * *noname*: refer, master h1. 6 Presence The requirements of the seminar are: * AS.BDA.RQ.1) deliver presentation: (i) on the topic of master thesis, (ii) review of a recognized paper. * AS.BDA.RQ.2) attend not less than 50% of classes. Here is a table of [[Presence]] conducted from meeting minutes (see minutes as issues in first column of the [[Seminars schedule]].