-
Machine Learning Driven Global Optimisation Framework for Analog Circuit Design
Authors:
Ria Rashid,
Komala Krishna,
Clint Pazhayidam George,
Nandakumar Nambath
Abstract:
We propose a machine learning-driven optimisation framework for analog circuit design in this paper. The primary objective is to determine the device sizes for the optimal performance of analog circuits for a given set of specifications. Our methodology entails employing machine learning models and spice simulations to direct the optimisation algorithm towards achieving the optimal design for anal…
▽ More
We propose a machine learning-driven optimisation framework for analog circuit design in this paper. The primary objective is to determine the device sizes for the optimal performance of analog circuits for a given set of specifications. Our methodology entails employing machine learning models and spice simulations to direct the optimisation algorithm towards achieving the optimal design for analog circuits. Machine learning based global offline surrogate models, with the circuit design parameters as the input, are built in the design space for the analog circuits under study and is used to guide the optimisation algorithm, resulting in faster convergence and a reduced number of spice simulations. Multi-layer perceptron and random forest regressors are employed to predict the required design specifications of the analog circuit. Since the saturation condition of transistors is vital in the proper working of analog circuits, multi-layer perceptron classifiers are used to predict the saturation condition of each transistor in the circuit. The feasibility of the candidate solutions is verified using machine learning models before invoking spice simulations. We validate the proposed framework using three circuit topologies--a bandgap reference, a folded cascode operational amplifier, and a two-stage operational amplifier. The simulation results show better optimum values and lower standard deviations for fitness functions after convergence. Incorporating the machine learning-based predictions proposed in the optimisation method has resulted in the reduction of spice calls by 56%, 59%, and 83% when compared with standard approaches in the three test cases considered in the study.
△ Less
Submitted 14 October, 2024; v1 submitted 26 February, 2024;
originally announced April 2024.
-
Performance Evaluation of Evolutionary Algorithms for Analog Integrated Circuit Design Optimisation
Authors:
Ria Rashid,
Gopavaram Raghunath,
Vasant Badugu,
Nandakumar Nambath
Abstract:
An automated sizing approach for analog circuits using evolutionary algorithms is presented in this paper. A targeted search of the search space has been implemented using a particle generation function and a repair-bounds function that has resulted in faster convergence to the optimal solution. The algorithms are tuned and modified to converge to a better optimal solution with less standard devia…
▽ More
An automated sizing approach for analog circuits using evolutionary algorithms is presented in this paper. A targeted search of the search space has been implemented using a particle generation function and a repair-bounds function that has resulted in faster convergence to the optimal solution. The algorithms are tuned and modified to converge to a better optimal solution with less standard deviation for multiple runs compared to standard versions. Modified versions of the artificial bee colony optimisation algorithm, genetic algorithm, grey wolf optimisation algorithm, and particle swarm optimisation algorithm are tested and compared for the optimal sizing of two operational amplifier topologies. An extensive performance evaluation of all the modified algorithms showed that the modifications have resulted in consistent performance with improved convergence for all the algorithms. The implementation of parallel computation in the algorithms has reduced run time. Among the considered algorithms, the modified artificial bee colony optimisation algorithm gave the most optimal solution with consistent results across multiple runs.
△ Less
Submitted 18 October, 2023;
originally announced October 2023.
-
A Multi Constrained Transformer-BiLSTM Guided Network for Automated Sleep Stage Classification from Single-Channel EEG
Authors:
Farhan Sadik,
Md Tanvir Raihan,
Rifat Bin Rashid,
Minhjaur Rahman,
Sabit Md Abdal,
Shahed Ahmed,
Talha Ibn Mahmud
Abstract:
Sleep stage classification from electroencephalogram (EEG) is significant for the rapid evaluation of sleeping patterns and quality. A novel deep learning architecture, ``DenseRTSleep-II'', is proposed for automatic sleep scoring from single-channel EEG signals. The architecture utilizes the advantages of Convolutional Neural Network (CNN), transformer network, and Bidirectional Long Short Term Me…
▽ More
Sleep stage classification from electroencephalogram (EEG) is significant for the rapid evaluation of sleeping patterns and quality. A novel deep learning architecture, ``DenseRTSleep-II'', is proposed for automatic sleep scoring from single-channel EEG signals. The architecture utilizes the advantages of Convolutional Neural Network (CNN), transformer network, and Bidirectional Long Short Term Memory (BiLSTM) for effective sleep scoring. Moreover, with the addition of a weighted multi-loss scheme, this model is trained more implicitly for vigorous decision-making tasks. Thus, the model generates the most efficient result in the SleepEDFx dataset and outperforms different state-of-the-art (IIT-Net, DeepSleepNet) techniques by a large margin in terms of accuracy, precision, and F1-score.
△ Less
Submitted 19 September, 2023;
originally announced September 2023.
-
Mood Classification of Bangla Songs Based on Lyrics
Authors:
Maliha Mahajebin,
Mohammad Rifat Ahmmad Rashid,
Nafees Mansoor
Abstract:
Music can evoke various emotions, and with the advancement of technology, it has become more accessible to people. Bangla music, which portrays different human emotions, lacks sufficient research. The authors of this article aim to analyze Bangla songs and classify their moods based on the lyrics. To achieve this, this research has compiled a dataset of 4000 Bangla song lyrics, genres, and used Na…
▽ More
Music can evoke various emotions, and with the advancement of technology, it has become more accessible to people. Bangla music, which portrays different human emotions, lacks sufficient research. The authors of this article aim to analyze Bangla songs and classify their moods based on the lyrics. To achieve this, this research has compiled a dataset of 4000 Bangla song lyrics, genres, and used Natural Language Processing and the Bert Algorithm to analyze the data. Among the 4000 songs, 1513 songs are represented for the sad mood, 1362 for the romantic mood, 886 for happiness, and the rest 239 are classified as relaxation. By embedding the lyrics of the songs, the authors have classified the songs into four moods: Happy, Sad, Romantic, and Relaxed. This research is crucial as it enables a multi-class classification of songs' moods, making the music more relatable to people's emotions. The article presents the automated result of the four moods accurately derived from the song lyrics.
△ Less
Submitted 18 July, 2023;
originally announced July 2023.
-
Persistent Homology for Breast Tumor Classification using Mammogram Scans
Authors:
Aras Asaad,
Dashti Ali,
Taban Majeed,
Rasber Rashid
Abstract:
An Important tool in the field topological data analysis is known as persistent Homology (PH) which is used to encode abstract representation of the homology of data at different resolutions in the form of persistence diagram (PD). In this work we build more than one PD representation of a single image based on a landmark selection method, known as local binary patterns, that encode different type…
▽ More
An Important tool in the field topological data analysis is known as persistent Homology (PH) which is used to encode abstract representation of the homology of data at different resolutions in the form of persistence diagram (PD). In this work we build more than one PD representation of a single image based on a landmark selection method, known as local binary patterns, that encode different types of local textures from images. We employed different PD vectorizations using persistence landscapes, persistence images, persistence binning (Betti Curve) and statistics. We tested the effectiveness of proposed landmark based PH on two publicly available breast abnormality detection datasets using mammogram scans. Sensitivity of landmark based PH obtained is over 90% in both datasets for the detection of abnormal breast scans. Finally, experimental results give new insights on using different types of PD vectorizations which help in utilising PH in conjunction with machine learning classifiers.
△ Less
Submitted 11 July, 2022; v1 submitted 6 January, 2022;
originally announced January 2022.
-
Area Optimisation of Two Stage Miller Compensated Op-Amp in 65 nm Using Hybrid PSO
Authors:
Ria Rashid,
Nandakumar Nambath
Abstract:
Analog circuit design can be formulated as a non-linear constrained optimisation problem that can be solved using any suitable optimisation algorithms. Different optimisation techniques have been reported to reduce the design time of analog circuits. A hybrid particle swarm optimisation algorithm with linearly decreasing inertia weight for the optimisation of analog circuit design is proposed in t…
▽ More
Analog circuit design can be formulated as a non-linear constrained optimisation problem that can be solved using any suitable optimisation algorithms. Different optimisation techniques have been reported to reduce the design time of analog circuits. A hybrid particle swarm optimisation algorithm with linearly decreasing inertia weight for the optimisation of analog circuit design is proposed in this study. The proposed method is used to design a two-stage operational amplifier circuit with Miller compensation. The results show that the proposed optimisation method can substantially reduce the design time needed for analog circuits.
△ Less
Submitted 17 June, 2021;
originally announced June 2021.
