Browsing by Author "Muralidhar, Nikhil"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
- ML-Assisted Optimization of Securities LendingPrasad, Abhinav; Arunachalam, Prakash; Motamedi, Ali; Bhattacharya, Ranjeeta; Liu, Beibei; McCormick, Hays; Xu, Shengzhe; Muralidhar, Nikhil; Ramakrishnan, Naren (ACM, 2023-11-27)This paper presents an integrated methodology to forecast the direction and magnitude of movements of lending rates in security markets. We develop a sequence-to-sequence (seq2seq) modeling framework that integrates feature engineering, motif mining, and temporal prediction in a unified manner to perform forecasting at scale in real-time or near real-time.We have deployed this approach in a large custodial setting demonstrating scalability to a large number of equities as well as newly introduced IPO-based securities in highly volatile environments.
- Science Guided Machine Learning: Incorporating Scientific Domain Knowledge for Learning Under Data Paucity and Noisy ContextsMuralidhar, Nikhil (Virginia Tech, 2022-08-18)In recent years, the large amount of labeled data available has helped tend machine learning (ML) research toward using purely data driven end-to-end pipelines, e.g., in deep neural network research. However, in many situations, data is limited and of poor quality. Traditional ML pipelines are known to be susceptible to various issues when trained on low volumes of non-representative, noisy datasets. We investigate the question of whether prior domain knowledge about the problem being modeled can be employed within the ML pipeline to improve model performance under data paucity and in noisy contexts? This report presents recent developments as well as details, novel contributions in the context of incorporating prior domain knowledge in various data-driven modeling (i.e., machine learning - ML) pipelines particularly geared towards scientific applications. Such domain knowledge exists in various forms and can be incorporated into the machine learning pipeline using different implicit and explicit methods (termed: science-guided machine learning (SGML)). All the novel techniques proposed in this report have been presented in the context of developing SGML to model fluid dynamics applications, but can be easily generalized to other applications. Specifically, we present SGML pipelines to (i) incorporate prior domain knowledge into the ML model architecture (ii) incorporate knowledge about the distribution of the target process as statistical priors for improved prediction performance (iii) leverage prior knowledge to quantify consistency of ML decisions with scientific principles (iv) explicitly incorporate known mathematical relationships of scientific phenomena to influence the ML pipeline (v) develop science-guided transfer learning to improve performance under data paucity. Each technique that is presented, has been designed with a focus on simplicity and minimal cost of implementation with a goal of yielding significant improvements in model performance especially under low data volumes or under noisy data conditions. In each application, we demonstrate through rigorous qualitative and quantitative experiments that our SGML pipelines achieve significant improvements in performance and interpretability over corresponding models that are purely data-driven and agnostic to scientific knowledge.
- Segmentations with Explanations for Outage AnalysisChen, Liangzhe; Muralidhar, Nikhil; Chinthavali, Supriya; Ramakrishnan, Naren; Prakash, B. Aditya (Department of Computer Science, Virginia Polytechnic Institute & State University, 2018-04-09)Recent hurricane events have caused unprecedented amounts of damage and severely threatened our public safety and economy. The most observable (and severe) impact of these hurricanes is the loss of electric power in many regions, which causes the breakdown of many public services. Understanding the power outages and how they evolve during a hurricane provide insights on how to reduce outages in the future, and how to improve the robustness of the underlying critical infrastructure systems. In this paper, we propose a novel segmentation with explanations framework to help experts understand such datasets. Our method, CUT-n-REVEAL, first finds a segmentation of the outage sequences to capture pattern changes in the sequences. We then propose a novel explanation optimization problem to find an intuitive explanation of the segmentation, that highlights the culprit of the change. Via extensive experiments, we show that our method performs consistently in multiple datasets with ground truth. We further study real county-level power outage data from several recent hurricanes (Matthew, Harvey, Irma) and show that CUT-n-REVEAL recovers important, nontrivial and actionable patterns for domain experts.
- Upward Bound and Talent Search VideoSrivastava, Arunima; Muralidhar, Nikhil (2012-05-03)Upward Bound and Educational Talent Search are both national programs housed at Virginia Tech to encourage and assist high school students in college admission preparation. This project was assigned with the goal of updating and revamping their very first recruiting video for current high school students.