Hirak J. Kashyap

Adarsha Balaji, Prathyusha Adiraju, Hirak J. Kashyap et al.

We present PyCARL, a PyNN-based common Python programming interface for hardware-software co-simulation of spiking neural network (SNN). Through PyCARL, we make the following two key contributions. First, we provide an interface of PyNN to CARLsim, a computationally-efficient, GPU-accelerated and biophysically-detailed SNN simulator. PyCARL facilitates joint development of machine learning models and code sharing between CARLsim and PyNN users, promoting an integrated and larger neuromorphic community. Second, we integrate cycle-accurate models of state-of-the-art neuromorphic hardware such as TrueNorth, Loihi, and DynapSE in PyCARL, to accurately model hardware latencies that delay spikes between communicating neurons and degrade performance. PyCARL allows users to analyze and optimize the performance difference between software-only simulation and hardware-software co-simulation of their machine learning models. We show that system designers can also use PyCARL to perform design-space exploration early in the product development stage, facilitating faster time-to-deployment of neuromorphic products. We evaluate the memory usage and simulation time of PyCARL using functionality tests, synthetic SNNs, and realistic applications. Our results demonstrate that for large SNNs, PyCARL does not lead to any significant overhead compared to CARLsim. We also use PyCARL to analyze these SNNs for a state-of-the-art neuromorphic hardware and demonstrate a significant performance deviation from software-only simulations. PyCARL allows to evaluate and minimize such differences early during model development.

Hirak J Kashyap, Charless Fowlkes, Jeffrey L Krichmar

Dynamic scenes that contain both object motion and egomotion are a challenge for monocular visual odometry (VO). Another issue with monocular VO is the scale ambiguity, i.e. these methods cannot estimate scene depth and camera motion in real scale. Here, we propose a learning based approach to predict camera motion parameters directly from optic flow, by marginalizing depthmap variations and outliers. This is achieved by learning a sparse overcomplete basis set of egomotion in an autoencoder network, which is able to eliminate irrelevant components of optic flow for the task of camera parameter or motionfield estimation. The model is trained using a sparsity regularizer and a supervised egomotion loss, and achieves the state-of-the-art performances on trajectory prediction and camera rotation prediction tasks on KITTI and Virtual KITTI datasets, respectively. The sparse latent space egomotion representation learned by the model is robust and requires only 5% of the hidden layer neurons to maintain the best trajectory prediction accuracy on KITTI dataset. Additionally, in presence of depth information, the proposed method demonstrates faithful object velocity prediction for wide range of object sizes and speeds by global compensation of predicted egomotion and a divisive normalization procedure.

Hirak Kashyap, Hasin Afzal Ahmed, Nazrul Hoque et al.

Bioinformatics research is characterized by voluminous and incremental datasets and complex data analytics methods. The machine learning methods used in bioinformatics are iterative and parallel. These methods can be scaled to handle big data using the distributed and parallel computing technologies. Usually big data tools perform computation in batch-mode and are not optimized for iterative processing and high data dependency among operations. In the recent years, parallel, incremental, and multi-view machine learning algorithms have been proposed. Similarly, graph-based architectures and in-memory big data tools have been developed to minimize I/O cost and optimize iterative processing. However, there lack standard big data architectures and tools for many important bioinformatics problems, such as fast construction of co-expression and regulatory networks and salient module identification, detection of complexes over growing protein-protein interaction data, fast analysis of massive DNA, RNA, and protein sequence data, and fast querying on incremental and heterogeneous disease networks. This paper addresses the issues and challenges posed by several big data problems in bioinformatics, and gives an overview of the state of the art and the future research opportunities.