Artificial Intelligence

NK Labs has expertise in artificial intelligence, machine learning and data science. Given a dataset, we use visualization tools (pandas, seaborn, matplotlib) before building models that utilize both supervised and unsupervised machine learning techniques (Scikit-learn). We have also implemented both shallow and deep neural networks for applications that utilize large sets of data  in Tensorflow.

 

Capabilities:

 

  • Feature selection

  • Supervised - Linear/Logistic Regressions, Support Vector Machines (SVMs)

  • Unsupervised - K-Means, Principal Component Analysis (PCA)

  • Algorithm development

 

Deep Learning with Artificial Neural Networks:

 

  • Hyperparameter tuning and Optimization

  • Convolutional Neural Networks (CNNs) as applied to a variety of image, video, and other 2D or 3D data

  • Recurrent Neural Networks (RNNs) & Long short-term memory (LSTMs) as applied to audio applications (NLP)

We can help with your project.

Gradient Descent algorithm with training rate "ɑ", iteration number "k", model feature "m", and output "d"

Electric capacitive tomography using 12 electrodes to reconstruct an image of two locations of water. Iterations of the reconstruction algorithm can be seen on the right converging on the original image.