Kishore Konda

I am a PostDoc researcher at center for cognition and computation, department of computer science, Goethe University, Frankfurt. My research interests include relational feature learning, deep learning for vision. Currently my focus is on  “Understanding the interplay between sparsity, gating connections, and linear encoding in neural networks.”

Contact

Kishore Konda
Department of computer science,
Goethe-Universität Frankfurt,
Robert-Mayer-Straße 10, D-60325 Frankfurt am Main.
email: konda[at]cs[dot]uni[dash]frankfurt[dot]de
Phone: +49-69798-25564

Publications

2016

  • 2016 Lin, Z., Memisevic, R., & Konda, K.
    How far can we go without convolution: Improving fully-connected networks.
    International Conference on Learning Representations (ICLR2016-Workshop).
    [pdf]

2015

  • 2015 Konda,K., Bouthillier, X., Memisevic, R., Vincent, P.
    Dropout as data augmentation
    arXiv:1506.08700
    [pdf]
  • 2015 Konda, K., Memisevic, R., Krueger, D.
    Zero-bias autoencoders and the benefits of co-adapting features
    International Conference on Learning Representations (ICLR 2015)
    [pdf]
  • 2015 Kahou, S. E., Michalski, V., Konda, K., Memisevic, R., Pal, C.,
    Recurrent Neural Networks for Emotion Recognition in Video
    ICMI 2015 (to appear).
  • 2015 Konda, K., Memisevic, R.
    Learning visual odometry with a convolutional network
    International Conference on Computer Vision Theory and Applications (VISAPP2015)
    [pdf]
  • 2015 Konda, K., Chandrasekariah, P., Memisevic, R., Triesh, J.
    Real-time activity recognition via deep learning of motion features
    European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN2015)
  • 2015 Kahou, S. E., Bouthillier, X., Lamblin, P., Gulcehre, C., Michalski, V., Konda, K., … , Jean, S., Froumenty, P., Courville, A., Vincent, P., Memisevic, R., Pal, C., Bengio, Y.
    EmoNets: Multimodal deep learning approaches for emotion recognition in video,
    Journal on Multimodal User Interfaces
    [pdf] [preprint]

2014

  • 2014 Konda, K., Memisevic, R.
    A unified approach to learning depth and motion features
    Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP2014)
    [pdf]
  • 2014 Michalski, V., Memisevic, R., Konda, K.
    Modeling Deep Temporal Dependencies with “Grammar Cells”
    Neural Information Processing Systems (NIPS 2014)
    [pdf][supplementary][webpage][bibtex]
  • 2014 Konda, K., Memisevic, R., Michalski, V.
    Learning to encode motion using spatio-temporal synchrony
    International Conference on Learning Representations (ICLR 2014)
    [pdf][bibtex]
  • 2014 Konda, K., Memisevic, R
    Learning visual odometry with a convolutional network
    Bernstein Conference 2014.

2013

  • 2013 Kahou, S. E., Pal, C., Bouthillier, X., Konda, K., *, Memisevic, R., Vincent, P., Courville, A. and Bengio, Y.
    Combining Modality Specific Deep Neural Networks for Emotion Recognition in Video
    ICMI 2013
    * see paper for additional authors. The paper describes our model that won the ICMI 2013 Grand Challenge on Emotion Recognition in the Wild.
    [pdf]
  • 2013 Konda K, Memisevic R, Michalski V
    Boltzmann machines with dendritic gating.
    Bernstein Conference 2013. doi: 10.12751/nncn.bc2013.0160

2012

  • 2012 Konda, K. R., Königs, A., Schulz, H., & Schulz, D.
    Real time interaction with mobile robots using hand gestures.
    In Proceedings of the seventh annual ACM/IEEE international conference on Human-Robot Interaction (pp.177-178). ACM.
    [pdf]