arXiv subjects

Ashutosh Jha

Publications and source records attributed to Ashutosh Jha.

4 recordsLinked to original sources

arXiv2026

Linear Independent Component Analysis via Optimal Transport

Linear Independent Component Analysis (ICA) recovers jointly independent source signals from their linear mixtures. To achieve this, classical ICA algorithms attempt to maximize non-Gaussianity, measured by negentropy, which is linked to independence by information theory. Because exact negentropy optimization is intractable, they rely on proxy contrast functions, such as fourth-order cumulants, and parametric log-likelihoods. We propose instead to measure non-Gaussianity using the squared Wasserstein distance $W_2^2$ to a standard Gaussian. We prove that the Wasserstein distance between a standard normal distribution and linear projections of the data is maximized when the projection recovers an independent component. Based on this observation, we propose the OT-ICA algorithm which finds this projection by gradient-based optimization. Empirical evaluation on simulated data shows that OT-ICA outperforms proxy-based methods for different distributions of the latent variables. Application to EEG artifact removal and econometric price discovery confirm OT-ICA can be used for applied ICA tasks without distributional assumptions.

cs.LG↗

arXiv2021

Supportive 5G Infrastructure Policies are Essential for Universal 6G: Assessment using an Open-source Techno-economic Simulation Model utilizing Remote Sensing

Work has now begun on the sixth generation of cellular technologies (`6G`) and cost-efficient global broadband coverage is already becoming a key pillar. Indeed, we are still far from providing universal and affordable broadband connectivity, despite this being a key part of the Sustainable Development Goals (Target 9.c). Currently, both Mobile Network Operators and governments still lack independent analysis of the strategies that can help achieve this target with the cellular technologies available (4G and 5G). Therefore, this paper undertakes quantitative assessment demonstrating how current 5G policies affect universal broadband, as well as drawing conclusions over how decisions made now affect future evolution to 6G. Using a method based on an open-source techno-economic codebase, combining remote sensing with least-cost network algorithms, performance analytics are provided for different 4G and 5G universal broadband strategies. As an example, the assessment approach is applied to India, the world`s second-largest mobile market and a country with very high spectrum prices. The results demonstrate the trade-offs between technological decisions. This includes demonstrating how important current infrastructure policy is, particularly given fiber backhaul will be essential for delivering 6G quality of service. We find that by eliminating the spectrum licensing costs, 100% 5G population coverage can viably be achieved using fiber backhaul. Therefore, supportive infrastructure policies are essential in providing a superior foundation for evolution to future cellular generation, such as 6G.

econ.GN↗

arXiv2018

Improvements on Hindsight Learning

Sparse reward problems are one of the biggest challenges in Reinforcement Learning. Goal-directed tasks are one such sparse reward problems where a reward signal is received only when the goal is reached. One promising way to train an agent to perform goal-directed tasks is to use Hindsight Learning approaches. In these approaches, even when an agent fails to reach the desired goal, the agent learns to reach the goal it achieved instead. Doing this over multiple trajectories while generalizing the policy learned from the achieved goals, the agent learns a goal conditioned policy to reach any goal. One such approach is Hindsight Experience replay which uses an off-policy Reinforcement Learning algorithm to learn a goal conditioned policy. In this approach, a replay of the past transitions happens in a uniformly random fashion. Another approach is to use a Hindsight version of the policy gradients to directly learn a policy. In this work, we discuss different ways to replay past transitions to improve learning in hindsight experience replay focusing on prioritized variants in particular. Also, we implement the Hindsight Policy gradient methods to robotic tasks.

cs.LG↗

arXiv2017

Learning to Multi-Task by Active Sampling

One of the long-standing challenges in Artificial Intelligence for learning goal-directed behavior is to build a single agent which can solve multiple tasks. Recent progress in multi-task learning for goal-directed sequential problems has been in the form of distillation based learning wherein a student network learns from multiple task-specific expert networks by mimicking the task-specific policies of the expert networks. While such approaches offer a promising solution to the multi-task learning problem, they require supervision from large expert networks which require extensive data and computation time for training. In this work, we propose an efficient multi-task learning framework which solves multiple goal-directed tasks in an on-line setup without the need for expert supervision. Our work uses active learning principles to achieve multi-task learning by sampling the harder tasks more than the easier ones. We propose three distinct models under our active sampling framework. An adaptive method with extremely competitive multi-tasking performance. A UCB-based meta-learner which casts the problem of picking the next task to train on as a multi-armed bandit problem. A meta-learning method that casts the next-task picking problem as a full Reinforcement Learning problem and uses actor critic methods for optimizing the multi-tasking performance directly. We demonstrate results in the Atari 2600 domain on seven multi-tasking instances: three 6-task instances, one 8-task instance, two 12-task instances and one 21-task instance.

cs.NE↗