BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//EE - ECPv5.10.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:EE
X-ORIGINAL-URL:https://ee.iisc.ac.in
X-WR-CALDESC:Events for EE
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
BEGIN:STANDARD
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
DTSTART:20230101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20230410T210000
DTEND;TZID=Asia/Kolkata:20230410T230000
DTSTAMP:20260528T222600
CREATED:20230410T000426Z
LAST-MODIFIED:20230410T001059Z
UID:240584-1681160400-1681167600@ee.iisc.ac.in
SUMMARY:Thesis colloquium of Mr. Anoop C. S.
DESCRIPTION:Advisor               : Prof. A. G. Ramakrishnan\n\nDate and Time: 10 April 2023 (Monday) 3:30 PM\n\n\n\nmeeting link: \n  \n\n\n\n\n\n\n\n\nJoin conversation\nteams.microsoft.com\n\n\n\n\n\n\n\nTITLE: Automatic speech recognition for low-resource Indian languages\n\nBuilding good models for automatic speech recognition (ASR) requires large amounts of annotated speech data. Most Indian languages are low-resourced and lack enough training data to build robust and efficient ASR systems. However\, many have an overlapping phoneme set and a strong correspondence between their character sets and pronunciations. In this thesis\, we exploit such similarities among the Indian languages to improve speech recognition in low-resource settings.\n\nSignificant contributions of the thesis:\n\nExploiting the pronunciation similarities across multiple Indian languages through shared label sets: \n\nWe propose the use of a common set of tokens across multiple Indian languages and analyze their performance in mono and multilingual settings.\n\n\nWe find that the Sanskrit Library Phonetic Encoding (SLP1) tokens\, which exploit the pronunciation-based structuring of character Unicodes in Indian languages\, perform better than some other grapheme-to-phoneme (G2P) based tokens in monolingual ASR settings.\nSyllable-based sub-words perform better than the character-based token units in monolingual speech recognition. However\, character-based SLP1 tokens perform better in cross-lingual transfer.\n\n\nStrategies for improving the performance of ASR systems in low-resource scenarios (target languages) exploiting the annotated data from high-resource languages (source languages):\n\nWe study three different low-resource settings:\n\nA) Labelled audio data is not available in the target language. Only a limited amount of unlabeled data is available. We adopt the unsupervised domain adaptation (UDA) schemes popular in image classification problems to tackle this case.\n\n\nThe adversarial training with gradient reversal layers (GRL) and domain separation networks (DSN) provides word error rate (WER) improvements of 6.71% and 7.32% in Sanskrit compared to a baseline hybrid DNN-HMM system trained on Hindi.\nThe UDA models outperform multi-task training with language recognition as the auxiliary task.\nSelection of the source language is critical in UDA systems.\n\n\nB) Target language has only a small amount of labeled data and has some amount of text data to build language models. We try to benefit from the available data in high-resource languages through shared label sets to build unified acoustic (AM) and language models (LM).\n\n\nUnified language-agnostic AM + LM performs better than monolingual AM + LM in cases where (a) only limited speech data is available for training the acoustic models and (b) the speech data is from domains different from that used in training.\nIn general\, multilingual AM + monolingual LM performs the best.\n\n\nC) There are N target languages with limited training data and several source languages with large training sets. Here\, we establish the usefulness of model-agnostic meta-learning (MAML) pre-training in Indian languages and propose improvements with text-similarity-based loss-weightings.\n\n\nMAML beats joint multilingual pretraining by an average of 5.4% in CER and 20.3% in WER.\nWith just 25% of the data\, MAML performance matches joint multilingual models trained on the whole target data.\nSimilarity with the source languages impacts the target language’s ASR performance.\nWe use text-similarity measured through cosine and Mahalanobis distances to weigh the losses during MAML pretraining. It yields a mean absolute improvement of 1% in WER.\n\n\n\n\n\n                                       ALL ARE WELCOME ONLINE!
URL:https://ee.iisc.ac.in/event/thesis-colloquium-of-mr-anoop-c-s/
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20230417T210000
DTEND;TZID=Asia/Kolkata:20230417T230000
DTSTAMP:20260528T222600
CREATED:20230409T224041Z
LAST-MODIFIED:20230412T233135Z
UID:240578-1681765200-1681772400@ee.iisc.ac.in
SUMMARY:Ph.D. Thesis colloquium of Ms. Ritika Jain
DESCRIPTION:Advisor: Prof. A. G. Ramakrishnan  \n\n\n   \n\n\nTITLE: Multimodal sleep staging and diagnosis of sleep disorders \n MS Teams link   \n\n\nSleep staging is a tedious and time-consuming process carried out manually by clinicians in which they annotate overnight polysomnograph recordings. An automated sleep scoring system can perform faster and objective sleep staging. Methods are proposed to classify the sleep EEG data into multiple stages by utilizing temporal\, spectral\, time-frequency\, non-linear\, and statistical features and random undersampling with boosting technique (RUSBoost) on a decision tree classifier. The role of data augmentation and temporal context on classifier performance is evaluated for healthy controls and clinical populations. This work also attempts to classify different sleep disorders using single-channel EEG and evaluate the role of individual sleep stages in that task.  \n\n\n  \n\n\nSignificant contributions of the thesis:         \n\n\n        \n\n\nBinary classification of sleep and wake states for healthy individuals and clinical population:   \n\n\n\nFor this two-class classification problem\, we explored the performances of different modalities such as EEG\, EOG & EMG.  \n\n\nWe also performed ensemble empirical mode decomposition and Poincare plot analysis of the signal for identifying sleep and wake states. \n\n\n\nMulti-class classification of sleep stages using single channel EEG:  \n\n\n\nUtilising the knowledge from earlier works on binary classification\, we considered different sets of features and evaluated the performance of RUSBoost classifier on unseen test subjects. This work reports the performance of different n-class (n=2\,3\,4\,5\,6) classification problems on three publicly available datasets of overnight polysomnography recordings. \n\n\n\nMulti-modal classification of sleep stages using a hierarchical model  \n\n\n\nIn this work\, a six-level hierarchical model (HM) has been designed. The aim is to improve the sleep staging accuracy by breaking down the 5-class classification problem into six binary classification problems\, while also reducing the misclassifications among N1\, REM\, and wake stages.  \n\n\nIntroducing data augmentation (DA) and temporal context (TC) in the proposed hierarchical model to further improve sleep staging performance. We validated the results of DA and TC on healthy as well as clinical populations from seven publicly available datasets. \n\n\n\nDiagnosis of different sleep disorders using a single EEG channel  \n\n\n\nThis work aims to classify seven different sleep disorders and healthy controls using light gradient boosting model with a single-channel EEG.   \n\n\nWe examined the importance of different features in distinguishing various pathological groups and healthy individuals.  \n\n\nWe also evaluated the role of individual sleep stages in distinguishing the different disorders. \n\n\n                                                                                                           ALL ARE WELCOME \nPublications based on this Thesis \n  \nJournals \n1. Ritika Jain and Angarai Ganesan Ramakrishnan. Electrophysiological and neuroimaging studies–during resting state and sensory stimulation in disorders of consciousness: a review. Frontiers Neurosc.\, 14:987\, 2020 \n2. Ritika Jain and Ramakrishnan A G. Reliable sleep staging of unseen subjects with fusion of multiple EEG features and RUSBoost. Biomed. Sig. Proc. Control\, 70:103061\, 2021 \n  \nConferences \n1. Ritika Jain and Ramakrishnan Angarai Ganesan. Assessment of submentalis muscle activity for sleep-wake classification of healthy individuals and patients with sleep disorders. \nIn 44th IEEE EMBC 2022. IEEE\, 2022 \n2. Ritika Jain and Ramakrishnan Angarai Ganesan. Single EOG channel performs well in distinguishing sleep from wake state for both healthy individuals and patients. In 44th \nIEEE EMBC. IEEE\, 2022 \n3. Ritika Jain and Ramakrishnan Angarai Ganesan. Poincar ́e plot analysis for sleep-wake classification of unseen patients using a single EEG channel. In 17th IEEE Int. Symp. \nMed. Meas. Applns. IEEE\, 2022 \n4. Ritika Jain and Ramakrishnan Angarai Ganesan. Classifying sleep-wake states of patients by training on single EEG or EOG channel data from normal subjects. In 2022 IEEE Region 10 Symposium (TENSYMP)\, pages 1–5. IEEE\, 2022 \n5. Ritika Jain and Ramakrishnan Angarai Ganesan. An efficient sleep scoring method using visibility graph and temporal features of single-channel EEG. In 43rd Ann. Int. Conf IEEE EMBC\, pages 6306–6309. IEEE\, 20216. Ramakrishnan A G and Ritika Jain. Binary state prediction of sleep or wakefulness using EEG and EOG features. In 17th India Council Int. Conf (INDICON)\, pages 1–7. IEEE\, 20207. Ritika Jain and Angarai Ganesan Ramakrishnan. Sleep-awake classification using EEG band-power-ratios and complexity measures. In 2020 IEEE 17th India Council International Conference (INDICON)\, pages 1–6. IEEE\, 2020Manuscripts under Review \n  \n• Ritika Jain and Ramakrishnan A G. Modality-specific feature selection\, data augmentation\, and temporal context for superior performance in sleep staging. IEEE Jl. Of Biomedical & Health Informatics\, 2023. \n  \n                                                                 ALL ARE WELCOME – People outside IISc can join through the MS Teams link given.
URL:https://ee.iisc.ac.in/event/ph-d-thesis-colloquium-of-ritika-jain/
LOCATION:MMCR\, Hall C 241\, 1st floor\, EE department
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20230424T160000
DTEND;TZID=Asia/Kolkata:20230424T180000
DTSTAMP:20260528T222600
CREATED:20230423T225041Z
LAST-MODIFIED:20230424T025205Z
UID:240653-1682352000-1682359200@ee.iisc.ac.in
SUMMARY:EE Seminar: High voltage aviation electrical system EMC challenges and opportunities
DESCRIPTION:Dear all\,\n\nI cordially invite you to attend this talk by Dr. Cong Li from GE Aerospace Research\, Niskayuna\, NY\, USA.  The details are given below.\n\nDate & Time: 24/04/2023\, 10.30 am – 11.30 am\n\nVenue: MMCR\, Dept. of Electrical Engineering (Hybrid Mode)\n\nFor online participants:Meeting Link\n\n\n\n\n\nTitle: High voltage aviation electrical system EMC challenges and opportunities\n\n\nAbstract: High voltage aviation electrical systems have unique design challenges to meet ultrahigh power density and reliability requirements under extreme operation conditions. One critical aspect is the Electromagnetic Compatibility (EMC)\, such as emission and susceptibility\, etc. This talk will explain the fundamental EMC challenges for wide band gap (WBG) based high voltage aviation electrical system\, and share an effective “SOLVE” EMC design process for meeting stringent aviation EMC requirements.\n\n\nSpeaker Bio: Dr. Cong Li (S’09-M’15-SM’19) received the Ph.D. degree in electrical engineering specializing in power electronics from The Ohio State University\, Columbus\, OH\, USA\, in 2014. He joined GE Aerospace Research at Niskayuna\, NY\, USA as a Research Engineer in 2014 and is currently a Senior Power Electronics Engineer and EMC Lead. His research interests include aviation high voltage\, high power\, high density wide band gap (WBG) power electronics systems\, and EMI mitigation techniques. He is currently leading multiple flight demo EMC projects at GE Research. He has authored more than dozens of technical papers\, and patent applications in the area of power electronics and EMC. He is a voting member of commercial aviation DO-160 EMI standard working group.\n\nHe has been given EMI webinars and tutorials at multiple IEEE conferences such as ECCE\, APEC\, EMC Symposium\, etc. He is currently a Senior Member of IEEE\, and Associate Editor at IEEE Open Journal of Power Electronics. He is serving as secretary of IEEE-EMCS-SC5 Power Electronics EMC\, as well as Technical Committee member of IEEE APEC\, ECCE\, ITEC\, EATS conferences.\n\n\nRegards\,\nVishnu
URL:https://ee.iisc.ac.in/event/ee-seminar-high-voltage-aviation-electrical-system-emc-challenges-and-opportunities/
LOCATION:MMCR\, Hall C 241\, 1st floor\, EE department
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20230424T203000
DTEND;TZID=Asia/Kolkata:20230424T223000
DTSTAMP:20260528T222600
CREATED:20230423T225448Z
LAST-MODIFIED:20230423T230059Z
UID:240656-1682368200-1682375400@ee.iisc.ac.in
SUMMARY:EE seminar: Advances in Power Electronics and Power Semiconductors for E-mobility Applications
DESCRIPTION:Dear all\, \n\nI cordially invite you to attend this talk by Dr. Ajay Poonjal Pai from Infineon Technologies AG\, Neubiberg\, Germany.  The details are given below.\n\nDate & Time: 24/04/2023\, 3 pm – 4 pm\n\nVenue: B 303 (Old 311)\, Dept. of Electrical Engineering (Hybrid Mode)\n\n\nFor online participants: Meeting Link\n  \nTitle of the Talk: Advances in Power Electronics and Power Semiconductors for E-mobility Applications \n\n\nAbstract: E-mobility has emerged as an interesting application for power electronics and power semiconductors. Not only is this market demonstrating an exponential growth\, but is also filled with interesting challenges. This is where Power semiconductors and power electronics can pitch in to facilitate adoption of electric cars. This talk gives an overview of the most interesting power electronic applications in electric Vehicles\, with a main focus on the traction inverter application\, which is\, by far\, the most important from a power semiconductor point of view. The latest trends in the application and power semiconductor technologies is discussed. Special focus will be given to Silicon Carbide Mosfet technology which offers significant benefits in terms of power density\, switching behaviour\, conduction behaviour etc. Results of power loss measurements are shown and discussed.\n\n\nSpeaker Bio: Dr. Ajay Poonjal Pai obtained his B.Tech in Electrical Engineering from NITK Surathkal\, India and M.Sc. in Electrical Power Engineering from RWTH Aachen University\, Germany. He then pursued his PhD focusing on Silicon Carbide at the Friedrich Alexander University\, Erlangen-Nuremberg\, Germany. Since 2015\, he is working at Infineon Technologies AG\, Neubiberg\, Germany as a Principal Application Engineer responsible for next-generation Silicon Carbide technologies and Power Modules for electric vehicles. His research interests include e-mobility\, Silicon Carbide semiconductors and power electronics. He enjoys following new technologies and understanding trends in the power electronics and automotive markets. He has contributed to several conferences and journals\, and has delivered numerous lectures and tutorials around the world. \n\n\n\n\nRegards\,\nVishnu
URL:https://ee.iisc.ac.in/event/ee-seminar-advances-in-power-electronics-and-power-semiconductors-for-e-mobility-applications/
LOCATION:B 303 (Old 311)\, Dept. of Electrical Engineering (Hybrid Mode)
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20230426T163000
DTEND;TZID=Asia/Kolkata:20230426T183000
DTSTAMP:20260528T222600
CREATED:20230425T224859Z
LAST-MODIFIED:20230425T224859Z
UID:240704-1682526600-1682533800@ee.iisc.ac.in
SUMMARY:IISc-TU Delft Talk by Prof Marjan Popov 26th April 11 am
DESCRIPTION:Dear All\, \nProf Marjan Popov of TU Delft will be visiting IISc on 26th April 2023. He will be giving a talk in the MMCR of the Electrical Engineering Department\, from 11 am on “Power System Protection Essentials – research activities”. He will mostly give an overview of the research activity carried out by his lab. This talk is a part of an IISc-TU Delft joint project. \nAll are invited.
URL:https://ee.iisc.ac.in/event/iisc-tu-delft-talk-by-prof-marjan-popov-26th-april-11-am/
LOCATION:MMCR\, Hall C 241\, 1st floor\, EE department
END:VEVENT
END:VCALENDAR