Click to listen highlighted text!

Indian Institute of Information Technology, Lucknow
भारतीय सूचना प्रौद्योगिकी संस्थान, लखनऊ
(An Institute of National Importance by the Act of Parliament)

Dr. Kanishka Tyagi

kanishka@iiitl.ac.in

Dr. Kanishka Tyagi

Visiting Faculty

EDUCATION

  • Ph.D. (The University of Texas at Arlington)
  • M.Sc. (The University of Texas at Arlington)
  • B.Tech ( G.B. Pant University of Agriculture & Technology, Pantnagar, India. )

RESEARCH INTEREST

  • Machine learning & neural networks
  • Optimization theory
  • Statistical signal processing
  • Automotive radar systems

PATENTS

  • N.Kumar, M.G.Garcia, K. Tyagi, “Material sorting using a vision system”, UHV technology, FortWorth, Texas, US20180243800A1., US Patent App. 15/963,755, 2018
  • N.Kumar, M.G.Garcia, K. Tyagi, “Material handling using machine leaning system”, Sortera alloys, Inc, FortWorth, Texas. (patent filed in July 2021)
  • X.Cai, D.K. Singh, B. Zhou, K. Tyagi, R. Sriram, “E-Commerce Platform and Marketplace for Exchange of IP’s related to Scientific Research”, (patent filing in December 2021).
  • K.Tyagi, J.Kirkwood. “Classification of Moving Objects with Low Level Radar Data”, Aptiv Corp, Agoura Hills, California. (patent filed in February 2020)
  • K.Tyagi, Y.Zhang, S. Zhang, S. Song, J. Kirkwood, N.Manukian, “Radar system using a machine learning model for stationary object detection”, Aptiv Corp, Agoura Hills, California. (patent filed in January 2021)
  • K.Tyagi, J. Kirkwood, N. Manukian, “Lower bound neural network training for stationary object detection using low level radar data”, Aptiv Corp, Agoura Hills, California. (patent filed in September 2021)
  • R.Billapati, K.Tyagi, N. Manukian, “Artificial intelligence on Radar Sensors to increase the accuracy of future predictions in radar machine learning using V2V technology”, Aptiv Corp, Agoura Hills, California and Kokomo Technical Center, Indiana. (provisional patent filed in July 2021)
  • K.Tyagi, K. Ahmadi, Y. Zhang, N. Manukian, “Super-resolution of Low level Radar Data using Neural Nets”. Aptiv Corp, Agoura Hills, California.(provisional patent filed in November 2021)
  • Y.Zhang, K. Tyagi, N.Manukian, “Enhanced Super-resolution of Low level Radar Data using Neural Nets”. Aptiv Corp, Agoura Hills, California. (provisional patent filed in November 2021)
  • Y.Zhang, K.Tyagi, N.Manukian, “Fuzzy labelling for reducing false positives in radar systems for autonomous driving”. Aptiv Corp, Agoura Hills, California. (provisional patent filed in January 2022)
  • R.Billapati, K. Tyagi, “Automated parking using mesh networks”, Aptiv Corp, Agoura Hills, California and Kokomo Technical Center, Indiana. (provisional patent filed in July 2021)

TRADE SECRETS

  • Jokanovic, W. Snyder K.Tyagi, Generative framework for neural net models using radar subsampling”. Aptiv Corp, Agoura Hills, California.

WORK EXPERIENCE

Aptiv Corporation, Agoura Hills, CA (April 2018- Current), Lead Machine Learning Autonomous Driving Scientist.
  • Algorithm development and hardware implementation of stationary object detection using low level radar
  • Algorithm development for machine learning based super-resolution
  • Algorithm development for moving object classification using low level radar
  • Algorithm development for fuzzy labelling of targets and sequence modelling for stationary
  • Identifying and infusing machine learning techniques in various radar systems for autonomous
  • Development of low-level radar data visualization tool for auto-labelling and internal data storage

 

 Google Research, Mountain View, CA (May 2017- August 2017), Summer Research Intern

  • Developed smaller and efficient neural networks for mobile vision using mixture of
  • Presented the research findings at Ph.D. research intern

 

·   The Mathworks, Natick, MA (May 2015 – August 2015) Machine Learning Development Intern
  • Developed distance metrics for heterogeneous mixtures for categorical & numerical data in C++ & Intel-TBB.
·     Siemens Energy Inc, Richland, MS (Aug 2011- Aug 2012), Technical Staff Member
  • Developing embedded based communication layer (ARM Processor) and firmware code for new product line on next generation control panel for step voltage Designed a frequency detection algorithm to be used in regulators.
·     Verizon Wireless, Irving, TX (Dec 2010- April 2011), Software Developer Intern
    • Designed Web Service interface using XML, XDS’s, Axis and Analyzed functional requirement and worked in the development of recommendation engine algorithms for FiOS project modules.

PUBLICATIONS

  • Conference proceedings:
    • Kheirkhah, K.Tyagi, S. Nguyen, M.Manry, “Structural adaptation for sparsely connected MLP using Newton’s method”, IEEE IJCNN’17, May 14-19, 2017, Anchorage, AK, USA.
    • Hao, K. Tyagi, R. Rawat, M. Manry, “Second order design of multiclass kernel machines”, IEEE IJCNN’16, July 24-29, 2016, Vancouver, BC, Canada.
    • Nguyen, K.Tyagi, P.Kheirkhah, M. Manry, “Partially affine invariant back propagation”, IEEE IJCNN’16, July 24-29, 2016, Vancouver, BC, Canada.
    • S. Auddy, K. Tyagi, S. Nguyen, M. Manry, “Discriminant vector transformations in neural network classifiers”, IEEE IJCNN’16, July 24-29, 2016, Vancouver, BC, Canada.
    • Cai, K. Tyagi, M. Manry, Z. Chen, “An efficient conjugate gradient-based learning algorithm for multiple optimal learning factors of multilayer perceptron neural network”, IEEE IJCNN’14, July 6-11, 2014, Beijing, China.
    • Tyagi, N.Kwak, M.Manry, “Optimal conjugate gradient algorithm for generalization of linear discriminant analysis based on L1 Norm”, ICPRAM’14, March, 2014, Loire Valley, France.
    • Godbole, K. Tyagi, M. Manry, “Neural decision directed segmentation of silicon defects”, IEEE IJCNN’13, Aug 4-9, 2013, Dallas, TX, USA.
    • Tyagi, X. Cai, M. Manry, “An optimal construction and training algorithm for radial basis neural network based on second order algorithm”, IEEE IJCNN’11, July 31-Aug 5, 2011, San Jose, CA, USA.
    • Tyagi, X. Cai, M. Manry, “Fuzzy C-Means clustering based construction and training algorithm for second order RBF network”, Oral presentation, IEEE Intl. Conf. on Fuzzy Systems, June 27-31, 2011, Taipei, Taiwan. (Best paper travel grant award )

    • Cai, K. Tyagi, M. Manry, “Training multilayer perceptron by using optimal input normalization”, Oral presentation, IEEE Intl. Conf. on Fuzzy Systems, June 27-31, 2011, Taipei, Taiwan.
    • K. Yadav, K. Tyagi, B.Shah, P. K. Kalra, “An FFT and correlation based approach for engine condition monitoring,” IJcICT-2010, January 9-10, 2010, IIMT Bhubaneswar, India.
    • Tyagi, D.Mishra, P.K.Kalra, “A novel complex-valued counterpropagation network”, IEEE Symposium on Computational Intelligence and Data Mining, April1-5, 2007, Hawaii, USA. (Best paper travel grant award )
    • Tyagi, R.Jain, H.J.S.Prasad, “A novel complex neural approach for real time flood forecasting”, International Conference on Water and Flood Management, March 12-14, 2007, Dhaka, Bangladesh.
    • Tyagi, V.Jindal, V.Kumar, “A complex valued neuron model for landslide assessment, International Conference on Landslides & Engineered Slopes, June 30- July 4, 2007, Xi’an, China.
    Journal Article:
    • Tyagi, M. Manry, “Multi-step training of a generalized linear classifier”, Neural Process Lett (2019) 50: 1341
    • Tyagi, S.Nguyen, R.Rawat, M.Manry, “Second Order Training and Sizing for the Multilayer Perceptron”, Neural Process Letters (2019)
    • Bo, X. Cai, Z.Chen, K.Tyagi, Z.Li, “Second Order Newton’s Method for Training Radial Basis Function Neural Networks”, Journal of Computer Research and Development ,vol 52, Issue 7, 2015.
    • K.Yadav, K.Tyagi, B.Shah, P.K.Kalra, “Audio Signal Based Engine Condition Monitoring: A FFT and Correlation based Approach” published at IEEE Transactions on Instrumentation and Measurement Issue 12, 2011.
    • Tyagi, C.A. Rane, B.Irie, M.Manry, “Multistage Newton’s approach for training radial basis function neural network”, SN Computer Science, vol 2:366 (2021).
    Tech Reports:
    • Tyagi, K.Lee, “ Applications of Deep Learning Network on Audio and Music Problems”, IEEE Computational Intelligence Society Walter Karplus Summer Research Grant 2013.
    • Jeong, K.Tyagi, K.Lee, “An Efficient Paradigm for Audio Tag Classification Using Sparse Autoencoder and Multi-kernel SVM”. Music Information Retrieval Evaluation exchange 2013).
    Book Chapters:
    • Tyagi, C.A. Rane, M.Manry, “Supervised learning”, Artificial Intelligence and Machine Learning for Edge Computing, Elsevier, ( to be published : Early 2022)
    • Tyagi, C.A Rane, R.M. Sriram, M.Manry, “Unsupervised learning”, Artificial Intelligence and Machine Learning for Edge Computing, Elsevier, ( to be published : Early 2022)
    • Tyagi, H. Tyagi, C.A.Rane, M.Manry, “Regression analysis”, Artificial Intelligence and Machine Learning for Edge Computing, Elsevier, ( to be published : Early 2022).
    • Zeba, P. Suman, K. Tyagi, “Type of Blockchain”, Distributed Computing to Blockchain: Architecture, Technology, and Applications, Elsevier, ( to be published : Early 2022).
    ·      K. Tyagi, N. Tyagi, “Machine learning methods a predictive tool for solid waste treatment and recycling processes

    – Where are we heading to?”, Machine Learning for Recycling, Elsevier ( to be published in Early 2022 )

    • A.Alcalde, K.Tyagi, “Geometrical Methods In Quantum Physics”, Quantum Computing – A shift from Bits to Qubits, Elsevier (to be published : Early 2022)
    • A. Alcalde, K.Tyagi, “The Measurement Problem and Information Geometry”, Quantum Computing – A shift from Bits to Qubits, Elsevier (to be published : Early 2022)
    Preprints ( manuscripts in preparation, under review )
    • Tyagi, C.A. Rane, M. Manry, “ Exploring the Linearity in deep learning features using Autoencoders” .
    • A. Rane, K. Tyagi, S. Malalur, Y. Shinge, M. Manry, “Training feed-forward networks using optimal input gains” .
    • Tyagi, C.A. Rane, S. Nguyen, M. Manry, “Understanding activations and loss functions in training multi-layer perceptron” .
    • A. Rane, K. Tyagi, M. Manry, “ Adaptive activations in shallow Convolutional Neural Networks”.
    • Tyagi, C.A. Alcalde, “Lei-group theory for low level autonomous radar”.
Translate »
Skip to content Click to listen highlighted text!