Article Per Year (5 Year)
p-Index From 2021 - 2026
0.61
P-Index
This Author published in this journals
All Journal Scientific Journal of Engineering Research
Awasthi, Lalit
Unknown Affiliation
Engineering

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : Scientific Journal of Engineering Research

 1 
Robust Positive-Unlabeled Learning via Bounded Loss Functions under Label Noise Awasthi, Lalit; Danso, Eric
Scientific Journal of Engineering Research Vol. 1 No. 3 (2025): September
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v1i3.2025.314

Abstract

Positive-Unlabeled (PU) learning has become a pivotal tool in scenarios where only positive samples are labeled, and negative labels are unavailable. However, in practical applications, the labeled positive data often contains noise such as mislabeled or outlier instances that can severely degrade model performance. This issue is exacerbated using traditional surrogate loss functions, many of which are unbounded and overly sensitive to mislabeled examples. To address this limitation, we propose a robust PU learning framework that integrates bounded loss functions, including ramp loss and truncated logistic loss, into the non-negative risk estimation paradigm. Unlike conventional loss formulations that allow noisy samples to disproportionately influence training, our approach caps each instance’s contribution, thereby reducing the sensitivity to label noise. We mathematically reformulate the PU risk estimator using bounded surrogates and demonstrate that this formulation maintains risk consistency while offering improved noise tolerance. A detailed framework diagram and algorithmic description are provided, along with theoretical analysis that bounds the influence of corrupted labels. Extensive experiments are conducted on both synthetic and real-world datasets under varying noise levels. Our method consistently outperforms baseline models such as unbiased PU (uPU) and non-negative PU (nnPU) in terms of classification accuracy, area under the receiver operating characteristic curve (ROC AUC), and precision-recall area under the curve (PR AUC). The ramp loss variant exhibits particularly strong robustness without sacrificing optimization efficiency. These results demonstrate that incorporating bounded losses is a principled and effective strategy for enhancing the reliability of PU learning in noisy environments.
Hybrid Machine Learning Framework for Joint Prediction of Window Mean and Bit Error Rate in SC-LDPC Decoding Bibi, Tanzeela; Zhou, Hua; Akbar, Sana; Awasthi, Lalit
Scientific Journal of Engineering Research Vol. 2 No. 1 (2026): March
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i1.2026.364

Abstract

Modern low-latency communication systems increasingly rely on spatially coupled low-density parity-check (SC-LDPC) codes combined with windowed decoding (WD) to achieve high reliability with reduced latency and memory requirements. However, evaluating the intrinsic trade-off between decoding complexity and error performance typically measured by the average window iteration count (WMEAN) and bit error rate (BER) still depends on computationally intensive Monte Carlo simulations, which limits rapid system optimization and real-time design exploration. To address this limitation, this paper proposes a hybrid machine learning framework for the joint, non-iterative prediction of WMEAN and BER using a single set of code and channel parameters. A high-fidelity dataset is generated through extensive SC-LDPC windowed decoding simulations across varying window sizes, coupling lengths, and signal-to-noise ratio (SNR) conditions. Based on this dataset, a multi-output Random Forest Regressor is trained to exploit the shared underlying decoding dynamics that govern both computational complexity and decoding reliability. The proposed model achieves accurate simultaneous prediction of WMEAN and BER, demonstrating strong generalization performance while significantly reducing system evaluation time compared to conventional simulation-based approaches. Feature-importance analysis further reveals the dominant influence of channel quality and coupling structure on both decoding effort and error performance. These results indicate that the proposed framework provides an effective surrogate modeling tool for fast design-space exploration and informed performance–complexity trade-off analysis. The methodology enables practical optimization of high-throughput SC-LDPC decoders and supports the development of adaptive and resource-efficient communication systems.
Federated Temporal Graph Learning for Weakly Supervised Bearing Anomaly Detection Darlami, Khagendra; Awasthi, Lalit
Scientific Journal of Engineering Research Vol. 2 No. 2 (2026): June
Publisher : PT. Teknologi Futuristik Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.64539/sjer.v2i2.2026.413

Abstract

Industrial bearing health monitoring is hindered by four interrelated challenges: high class imbalance, the absence of fault-type annotations, stringent data privacy constraints prohibiting centralized aggregation, and non-independent and identically distributed (non-IID) degradation dynamics across geographically dispersed assets. To address these, we propose Fed-TGCN, a novel weakly supervised federated learning framework grounded in temporal graph neural networks. Each client represents a leave-one-bearing-out fold, comprising two training bearings, one validation bearing, and one held-out test bearing, constructs a hybrid spatio-temporal graph from six physics-informed statistical features derived from raw vibration signals; edges encode both sequential dependencies and feature-space similarity via k-nearest neighbors. Pseudo-anomaly labels are generated locally through adaptive thresholding of a degradation score using exponentially weighted moving average, eliminating reliance on expert annotations. Under a strict leave-one-bearing-out protocol on the NASA IMS dataset (12 bearings), local Temporal Graph Convolutional Networks are trained in isolation and aggregated globally via FedAvg. Our method achieves an Average Precision of 0.675 ± 0.276 and Matthews Correlation Coefficient of 0.636 ± 0.285, maintains stronger performance consistency across heterogeneous bearing conditions than isolated and non-graph baselines (ΔMCC = +0.130, p < 0.01). Ablation studies confirm the necessity of temporal modeling (MCC drops by 0.069 without GRU). To the best of our knowledge, this is the first work integrating weakly supervised, graph-based federated learning for bearing prognostics under, demonstrating that parameter coordination but not the data sharing which enables degradation-invariant representation learning across heterogeneous assets.
Co-Authors Akbar, Sana Bibi, Tanzeela Danso, Eric Darlami, Khagendra Zhou, Hua