Deep neural networks and their applications in robotics

• Název práce v češtině: Deep neural networks and their applications in robotics
• Název v anglickém jazyce: Deep neural networks and their applications in robotics
• Klíčová slova: asociativní paměti, vrstevnaté neuronové sítě, hluboké neuronové sítě, počítačové vidění, robotika
• Klíčová slova anglicky: associative memories, multilayer neural networks, deep neural networks, computer vision, robotics
• Akademický rok vypsání: 2016/2017
• Typ práce: diplomová práce
• Ústav: Katedra teoretické informatiky a matematické logiky (32-KTIML)
• Vedoucí / školitel: doc. RNDr. Iveta Mrázová, CSc.
• Řešitel: skrytý - zadáno a potvrzeno stud. odd.
• Datum přihlášení: 25.05.2017
• Datum zadání: 25.05.2017
• Datum potvrzení stud. oddělením: 13.06.2017

Zásady pro vypracování

The student shall review the following topics in his diploma thesis:

- overview and comparison of relevant (deep) neural network models (associative memories of the Hopfield and BAM type, multi-layered neural networks of the back-propagation type, convolutional neural networks)

- description of the main principles specific to the NAO humanoid robot platform

- recapitulation of known approaches suitable to high-performance implementation of artificial neural networks (GPU, C++/CUDA, etc.) and evaluation of their applicability for the NAO platform

- propose, discuss and test various aspects of the design and implementations of the investigated neural network models applicable to computer vision and robotics

The student will focus on some of these topics in more detail and will implement the discussed models. The evaluation of the obtained results and gained experience shall form an important part of the thesis.

Seznam odborné literatury

Seznam doporučené literatury:

1. Některé z dostupných základních učebnic, monografií či přehledových článků a publikací vhodných pro zvolené téma, např.:
- S. Haykin: Neural Networks and Learning Machines, 3rd edition, Pearson, 2009
- C.M. Bishop: Pattern Recognition and Machine Learning, Springer, 2006
- R. Rojas: Neural Networks: A Systematic Introduction, Springer, 1996
- M. Beiter, B. Coltin, S. Liemhetcharat: An Introduction To Robotics With NAO, Aldeberan Robotics, France, 2012
(accessible at: http://www.kramirez.net/Robotica/Material/Nao/AnIntroductionToRoboticsWithNao_TextBook_2012_US.pdf)
- A. Hughes: Working with the NAO Humanoid Robot, Florida Gulf Coast University, Fort Myers, Florida, 2013
(accessible at: http://itech.fgcu.edu/faculty/zalewski/projects/files/hughesworkingwithnaozv7.pdf)


2. Články:
- D. E. Rumelhart, G. E. Hinton, R. J. Williams: Learning representations by back-propagating errors, in: Nature, vol. 323, (1986) pp. 533-536.
- Y. LeCun, L. Bottou, Y. Bengio, P. Haffner: Gradient-Based Learning Applied to Document Recognition, in: Proc. of the IEEE, vol. 86, no. 11 (Nov. 1998), pp. 2278-2324.
- J. J. Hopfield: Neural networks and physical systems with emergent collective computational abilities, in: Proc. of the Nat. Acad. of Sciences, vol. 79, no. 8 (1982), pp. 2554-2558.
- D. Ciresan, U. Meier, J. Masci, J. Schmidhuber: A Committee of Neural Networks for Traffic Sign Competition, in: Proc. of IJCNN 2011, IEEE, (2011), pp. 1918-1921.
- K.-S. Oh, , K. Jung: GPU implementation of neural networks, in: Pattern Recognition, vol. 37, no. 6, (June 2004), pp. 1311–1314.
- G. I. Parisi, S. Wermter: Hierarchical SOM-based Detection of Novel Behavior for 3D Human Tracking, in: Proc. of IJCNN 2013, IEEE, (2013), pp. 1380-1387.
- K.-S. Oh, , K. Jung: GPU implementation of neural networks, in: Pattern Recognition, vol. 37, no. 6, (June 2004), pp. 1311–1314.


3. Aktuální články z profilujících světových časopisů, např.:
Neurocomputing, Neural Networks, IEEE Transactions on Neural Networks ap.