In this work we have described CMRNet, a CNN based approach for camera to LiDAR-Map registration, using the KITTI dataset for both learning and validation purposes. The performances of the proposed approach allow multiple specialized CMRNet to be st…

In this paper, we present an algorithm called best-scored clustering forest to efficiently solve the single-level density-based clustering problem. From the theoretical perspective, our main results comprise statements and complete analysis of stati…

We have concerned two main directions of universal model-free RL algorithm design and attempted to recreate several state-of-art pipelines.

In this paper, we developed a single-channel source separation model that uses additional conditioning on separate speaker context recordings. The model learns to create speaker embeddings for unseen speakers from additional context recordings. The …

We introduced a novel approach to latent tree learning: a relaxed version of stochastic differentiable dynamic programming which allows for efficient sampling of projective dependency trees and enables end-to-end differentiation. We demonstrate effe…

In this paper, we investigate a deep learning based generative model to estimate the ETA of a bus trip in real time. We train a model for each individual route using historical data of trips collected over two months. We observe that we could learn …

We have presented a κ-jet model for the accreting black hole in M 87 based on an AMR GRMHD simulation in Cartesian-Kerr-Schild coordinates, coupled to radiative-transfer calculations that include sub-grid models for electron acceleration based on re…

In this paper, we presented a hybrid-learning method to offer situation awareness utilizing situation templates and decision trees. We tested our method against a growing collection of data to simulate the real world example of smart environment’s a…

In this paper, we present an efficient kinodynamic replanning framework by exploiting the advantageous properties of the B-spline. The proposed EBK search algorithm is flexible and provides a user-specified parameter which can be used to control the…

In this work, a systematic review of emotionally-aware chatbots is proposed. We focus on three main issues, including, how to incorporate affective information into chatbots, what are resources that available and can be used to build EAC, and how to…

The authors declare no conflict of interest.

We present a new form of activation function which is motivated from polynomial approximation of univariate functions. The activation is learned during training while searching a space of finite degree polynomials. We provide in depth analysis of NN…

Machine Learning based Quality of Experience (QoE) models potentially suffer from over-fitting due to limitations including low data volume, and limited participant profiles. This prevents models from becoming generic. Consequently, these trained mo…

In situations where multi-class data are available, the objective of similarity learning can be naturally formulated as a ROC curve optimization problem, whose solutions are given by similarity functions yielding a maximal true positive rate with a …

We envision several further research questions on the topic of distributed tuplewise learning. We would like to provide a rigorous convergence rate analysis of the general distributed SGD algorithm introduced in Section 4. This is a challenging task…