arXiv subjects

Suryoday Basak

Publications and source records attributed to Suryoday Basak.

12 recordsLinked to original sources

arXiv2024

mmWave-Whisper: Phone Call Eavesdropping and Transcription Using Millimeter-Wave Radar

This paper introduces mmWave-Whisper, a system that demonstrates the feasibility of full-corpus automated speech recognition (ASR) on phone calls eavesdropped remotely using off-the-shelf frequency modulated continuous wave (FMCW) millimeter-wave radars. Operating in the 77-81 GHz range, mmWave-Whisper captures earpiece vibrations from smartphones, converts them into audio, and processes the audio to produce speech transcriptions automatically. Unlike previous work that focused on loudspeakers or limited vocabulary, this is the first work to perform such a speech recognition by handling large vocabulary and full sentences on earpiece vibrations from smartphones. This approach expands the potential of radar-audio eavesdropping. mmWave-Whisper addresses challenges such as the lack of large scale training datasets, low SNR, and limited frequency information in radar data through a systematic pipeline designed to leverage synthetic training data, domain adaptation, and inference by incorporating OpenAI's Whisper automatic speech recognition model. The system achieves a word accuracy rate of 44.74% and a character accuracy rate of 62.52% over a range of 25 cm to 125 cm. The paper highlights emerging misuse modalities of AI as the technology evolves rapidly.

cs.SD↗

arXiv2024

The ensmallen library for flexible numerical optimization

We overview the ensmallen numerical optimization library, which provides a flexible C++ framework for mathematical optimization of user-supplied objective functions. Many types of objective functions are supported, including general, differentiable, separable, constrained, and categorical. A diverse set of pre-built optimizers is provided, including Quasi-Newton optimizers and many variants of Stochastic Gradient Descent. The underlying framework facilitates the implementation of new optimizers. Optimization of an objective function typically requires supplying only one or two C++ functions. Custom behavior can be easily specified via callback functions. Empirical comparisons show that ensmallen outperforms other frameworks while providing more functionality. The library is available at https://ensmallen.org and is distributed under the permissive BSD license.

cs.MS↗

arXiv2023

Flexible numerical optimization with ensmallen

This report provides an introduction to the ensmallen numerical optimization library, as well as a deep dive into the technical details of how it works. The library provides a fast and flexible C++ framework for mathematical optimization of arbitrary user-supplied functions. A large set of pre-built optimizers is provided, including many variants of Stochastic Gradient Descent and Quasi-Newton optimizers. Several types of objective functions are supported, including differentiable, separable, constrained, and categorical objective functions. Implementation of a new optimizer requires only one method, while a new objective function requires typically only one or two C++ methods. Through internal use of C++ template metaprogramming, ensmallen provides support for arbitrary user-supplied callbacks and automatic inference of unsupplied methods without any runtime overhead. Empirical comparisons show that ensmallen outperforms other optimization frameworks (such as Julia and SciPy), sometimes by large margins. The library is available at https://ensmallen.org and is distributed under the permissive BSD license.

cs.MS↗

arXiv2023

mlpack 4: a fast, header-only C++ machine learning library

For over 15 years, the mlpack machine learning library has served as a "swiss army knife" for C++-based machine learning. Its efficient implementations of common and cutting-edge machine learning algorithms have been used in a wide variety of scientific and industrial applications. This paper overviews mlpack 4, a significant upgrade over its predecessor. The library has been significantly refactored and redesigned to facilitate an easier prototyping-to-deployment pipeline, including bindings to other languages (Python, Julia, R, Go, and the command line) that allow prototyping to be seamlessly performed in environments other than C++. mlpack is open-source software, distributed under the permissive 3-clause BSD license; it can be obtained at https://mlpack.org

cs.MS↗

arXiv2021

Quantifying the Classification of Exoplanets: in Search for the Right Habitability Metric

What is habitability? Can we quantify it? What do we mean under the term habitable or potentially habitable planet? With estimates of the number of planets in our Galaxy alone running into billions, possibly a number greater than the number of stars, it is high time to start characterizing them, sorting them into classes/types just like stars, to better understand their formation paths, their properties and, ultimately, their ability to beget or sustain life. After all, we do have life thriving on one of these billions of planets, why not on others? Which planets are better suited for life and which ones are definitely not worth spending expensive telescope time on? We need to find sort of quick assessment score, a metric, using which we can make a list of promising planets and dedicate our efforts to them. Exoplanetary habitability is a transdisciplinary subject integrating astrophysics, astrobiology, planetary science, even terrestrial environmental sciences. We review the existing metrics of habitability and the new classification schemes of extrasolar planets and provide an exposition of the use of computational intelligence techniques to evaluate habitability scores and to automate the process of classification of exoplanets. We examine how solving convex optimization techniques, as in computing new metrics such as CDHS and CEESA, cross-validates ML-based classification of exoplanets. Despite the recent criticism of exoplanetary habitability ranking, this field has to continue and evolve to use all available machinery of astroinformatics, artificial intelligence and machine learning. It might actually develop into a sort of same scale as stellar types in astronomy, to be used as a quick tool of screening exoplanets in important characteristics in search for potentially habitable planets for detailed follow-up targets.

astro-ph.EP↗

arXiv2018

SBAF: A New Activation Function for Artificial Neural Net based Habitability Classification

We explore the efficacy of using a novel activation function in Artificial Neural Networks (ANN) in characterizing exoplanets into different classes. We call this Saha-Bora Activation Function (SBAF) as the motivation is derived from long standing understanding of using advanced calculus in modeling habitability score of Exoplanets. The function is demonstrated to possess nice analytical properties and doesn't seem to suffer from local oscillation problems. The manuscript presents the analytical properties of the activation function and the architecture implemented on the function. Keywords: Astroinformatics, Machine Learning, Exoplanets, ANN, Activation Function.

cs.LG↗

arXiv2018

A Novel Exoplanetary Habitability Score via Particle Swarm Optimization of CES Production Functions

The search for life has two goals essentially: looking for planets with Earth-like conditions (Earth similarity) and looking for the possibility of life in some form (habitability). Determining habitability from exoplanet data requires that determining parameters are collectively considered before coming up with a conclusion as no single factor alone contributes to it. Our proposed models, would serve as an indicator while looking for new habitable worlds, if computed with precision and efficiency. The models are of the type constrained optimization, multivariate, convex but may suffer from curvature violation and premature convergence impacting desired habitability scores. We mitigate the problem by proposing modified Particle Swarm Optimization (PSO) to tackle constraints and ensuring global optima. In the process, a python library to tackle such problems has been created.

astro-ph.IM↗

arXiv2018

Habitability Classification of Exoplanets: A Machine Learning Insight

We explore the efficacy of machine learning (ML) in characterizing exoplanets into different classes. The source of the data used in this work is University of Puerto Rico's Planetary Habitability Laboratory's Exoplanets Catalog (PHL-EC). We perform a detailed analysis of the structure of the data and propose methods that can be used to effectively categorize new exoplanet samples. Our contributions are two fold. We elaborate on the results obtained by using ML algorithms by stating the accuracy of each method used and propose the best paradigm to automate the task of exoplanet classification. The exploration led to the development of new methods fundamental and relevant to the context of the problem and beyond. Data exploration and experimentation methods also result in the development of a general data methodology and a set of best practices which can be used for exploratory data analysis experiments.

astro-ph.IM↗

arXiv2018

A Comparative Analysis of the Cobb-Douglas Habitability Score (CDHS) with the Earth Similarity Index (ESI)

We present an analytical comparison of the Cobb-Douglas Habitability Production Function (CD-HPF) and the Earth Similarity Index (ESI). The key differences between the ESI and CD-HPF are highlighted and based on mathematical analysis, we show that the CD-HPF satisfies the conditions for model scalability and stability but the ESI does not. Using visualizations, we also demonstrate that there do not exist any causal relationships between the ESI and CD-HPF. The conclusion from the work done is that the CD-HPF and ESI do not share any sensible relationship and that both should be used independently.

astro-ph.IM↗

arXiv2018

Machine Learning in Astronomy: A Case Study in Quasar-Star Classification

We present the results of various automated classification methods, based on machine learning (ML), of objects from data releases 6 and 7 (DR6 and DR7) of the Sloan Digital Sky Survey (SDSS), primarily distinguishing stars from quasars. We provide a careful scrutiny of approaches available in the literature and have highlighted the pitfalls in those approaches based on the nature of data used for the study. The aim is to investigate the appropriateness of the application of certain ML methods. The manuscript argues convincingly in favor of the efficacy of asymmetric AdaBoost to classify photometric data. The paper presents a critical review of existing study and puts forward an application of asymmetric AdaBoost, as an offspring of that exercise.

astro-ph.IM↗

arXiv2018

Model Visualization in understanding rapid growth of a journal in an emerging area

A recent independent study resulted in a ranking system which ranked Astronomy and Computing (ASCOM) much higher than most of the older journals highlighting the niche prominence of the particular journal. We investigate the remarkable ascendancy in reputation of ASCOM by proposing a novel differential equation based modeling. The Modeling is a consequence of knowledge discovery from big data-centric methods, namely L1-SVD. The inadequacy of the ranking method in explaining the reason behind the growth in reputation of ASCOM is reasonable to understand given that the study was post-facto. Thus, we propose a growth model by accounting for the behavior of parameters that contribute to the growth of a field. It is worthwhile to spend some time in analysing the cause and control variables behind rapid rise in reputation of a journal in a niche area. We intent to probe and bring out parameters responsible for its growing influence. Delay differential equations are used to model the change of influence on a journal's status by exploiting the effects of historical data.

cs.DL↗

arXiv2017

Theoretical Validation of Potential Habitability via Analytical and Boosted Tree Methods: An Optimistic Study on Recently Discovered Exoplanets

Seven Earth-sized planets, TRAPPIST-1 system, were discovered in February 2017. Three of these planets are in the habitable zone (HZ) of their star, making them potentially habitable planets a mere 40 light years away. Discovery of the closest potentially habitable planet to us just a year before -- Proxima~b, and a realization that Earth-type planets in HZ are a common occurrence provides the impetus to the pursuit for life outside the Solar System. The search for life has two goals: Earth similarity and habitability. An index was recently proposed, Cobb-Douglas Habitability Score (CDHS), based on Cobb-Douglas production function, which computes the habitability score by using measured and estimated planetary parameters like radius, density, escape velocity and surface temperature of a planet. The proposed metric with exponents accounting for metric elasticity, is endowed with analytical properties that ensure global optima and can be scaled to accommodate a finite number of input parameters. We show that the model is elastic, and the conditions on elasticity to ensure global maxima can scale as the number of predictor parameters increase. K-Nearest Neighbour classification algorithm, embellished with probabilistic herding and thresholding restriction, utilizes CDHS scores and labels exoplanets to appropriate classes via feature-learning methods. The algorithm works on top of a decision-theoretical model using the power of convex optimization and machine learning. A second approach, based on a novel feature-learning and tree-building method classifies the same planets without computing the CDHS of the planets and produces a similar outcome. The convergence of the two different approaches indicates the strength of the proposed scheme and the likelihood of the potential habitability of the recent discoveries.

astro-ph.EP↗