Hamid Miah - Academia.edu (original) (raw)
Papers by Hamid Miah
arXiv Computer Science, Aug 21, 2020
ArXiv, 2021
Conventional domain generalization aims to learn domain invariant representation from multiple do... more Conventional domain generalization aims to learn domain invariant representation from multiple domains, which requires accurate annotations. In realistic application scenarios, however, it is too cumbersome or even infeasible to collect and annotate the large mass of data. Yet, web data provides a free lunch to access a huge amount of unlabeled data with rich style information that can be harnessed to augment domain generalization ability. In this paper, we introduce a novel task, termed as semi-supervised domain generalization, to study how to interact the labeled and unlabeled domains, and establish two benchmarks including a web-crawled dataset, which poses a novel yet realistic challenge to push the limits of existing technologies. To tackle this task, a straightforward solution is to propagate the class information from the labeled to the unlabeled domains via pseudo labeling in conjunction with domain confusion training. Considering narrowing domain gap can improve the quality...
ArXiv, 2021
It is a strong prerequisite to access source data freely in many existing unsupervised domain ada... more It is a strong prerequisite to access source data freely in many existing unsupervised domain adaptation approaches. However, source data is agnostic in many practical scenarios due to the constraints of expensive data transmission and data privacy protection. Usually, the given source domain pre-trained model is expected to optimize with only unlabeled target data, which is termed as source-free unsupervised domain adaptation. In this paper, we solve this problem from the perspective of noisy label learning, since the given pre-trained model can pre-generate noisy label for unlabeled target data via directly network inference. Under this problem modeling, incorporating self-supervised learning, we propose a novel Self-Supervised Noisy Label Learning method, which can effectively fine-tune the pre-trained model with pre-generated label as well as selfgenerated label on the fly. Extensive experiments had been conducted to validate its effectiveness. Our method can easily achieve stat...
Due to the character of the original source materials and the nature of batch digitization, quali... more Due to the character of the original source materials and the nature of batch digitization, quality control issues may be present in this document. Please report any quality issues you encounter to digital@library.tamu.edu, referencing the URI of the item.Not availabl
This report was prepared as a visioning and discussion piece for the CGIAR Challenge Program on W... more This report was prepared as a visioning and discussion piece for the CGIAR Challenge Program on Water and Food's Ganges Basin Development Challenge. The views expressed in this report are those of the consultant and in no way are representative of the thinking of the Challenge Program on Water and Food.
Pakistan Journal of Biological Sciences, Jun 1, 2000
One month old fry of a planktivorous Indian major carp species, catla, Catla catla (Hamilton) was... more One month old fry of a planktivorous Indian major carp species, catla, Catla catla (Hamilton) was reared upto fingerling stage in earthen ponds in order to determine the food preference and selection using Ivlev's Electivity Index under two different treatments. Except Chlorophyceae, it showed positive selection to Cyanophyceae, Euglenophyceae, Rodophyceae, Crustacea and Rotifera. However, Catla, during its fry-fingerling stage exerted strong preference for zooplankton and almost neutral response to phytoplankton in both conditions.
Pakistan Journal of Biological Sciences
Tropical Pest Management, 1986
Experiments were conducted to find out suitable pesticides, dose and method of application to con... more Experiments were conducted to find out suitable pesticides, dose and method of application to control white grubs (Scarabid beetles) damaging sugarcane in northern and northwestern districts of Bangladesh. Pesticides tested were Dieldrin (50WP, 40WP, 20EC), Heptachlor (40WP, 20G), Aldrin 5D, Ekalux 5G (Quinalphos) and Furadan 3G (Carbofuran) at varying doses and methods. Furadan 3G at 2.0 kg a.i./ha and 1.5
"Le Design des interfaces numériques est un ouvrage co-écrit par l'APCI... more "Le Design des interfaces numériques est un ouvrage co-écrit par l'APCI, *Designers Interactifs*, Mov'eo et Systematic Paris-Region. Les interfaces numériques sont au cœur du déploiement de dispositifs interactifs, dans différents domaines comme les transports, les communications, les services... Pour travailler à la conception de nouveaux outils et devices (interfaces Homme-Machine, applications logicielles, sites Web, objets connectés…), les professionnels de ces différents secteurs doivent apprendre à parler un seul et même langage. Cet ouvrage rassemble les définitions des 170 mots-clés du design des interfaces numériques, dans ses domaines fondamentaux que sont le design d’interaction, l'ingénierie, l’ergonomie, l’architecture de l’information. Composé par autant de termes du quotidien que de concepts spécialisés, cet ouvrage s'inscrit dans une démarche pédagogique, à destination des professionnels (designers, ingénieurs, ergonomes…) et des étudiants."
arXiv Computer Science, Aug 21, 2020
ArXiv, 2021
Conventional domain generalization aims to learn domain invariant representation from multiple do... more Conventional domain generalization aims to learn domain invariant representation from multiple domains, which requires accurate annotations. In realistic application scenarios, however, it is too cumbersome or even infeasible to collect and annotate the large mass of data. Yet, web data provides a free lunch to access a huge amount of unlabeled data with rich style information that can be harnessed to augment domain generalization ability. In this paper, we introduce a novel task, termed as semi-supervised domain generalization, to study how to interact the labeled and unlabeled domains, and establish two benchmarks including a web-crawled dataset, which poses a novel yet realistic challenge to push the limits of existing technologies. To tackle this task, a straightforward solution is to propagate the class information from the labeled to the unlabeled domains via pseudo labeling in conjunction with domain confusion training. Considering narrowing domain gap can improve the quality...
ArXiv, 2021
It is a strong prerequisite to access source data freely in many existing unsupervised domain ada... more It is a strong prerequisite to access source data freely in many existing unsupervised domain adaptation approaches. However, source data is agnostic in many practical scenarios due to the constraints of expensive data transmission and data privacy protection. Usually, the given source domain pre-trained model is expected to optimize with only unlabeled target data, which is termed as source-free unsupervised domain adaptation. In this paper, we solve this problem from the perspective of noisy label learning, since the given pre-trained model can pre-generate noisy label for unlabeled target data via directly network inference. Under this problem modeling, incorporating self-supervised learning, we propose a novel Self-Supervised Noisy Label Learning method, which can effectively fine-tune the pre-trained model with pre-generated label as well as selfgenerated label on the fly. Extensive experiments had been conducted to validate its effectiveness. Our method can easily achieve stat...
Due to the character of the original source materials and the nature of batch digitization, quali... more Due to the character of the original source materials and the nature of batch digitization, quality control issues may be present in this document. Please report any quality issues you encounter to digital@library.tamu.edu, referencing the URI of the item.Not availabl
This report was prepared as a visioning and discussion piece for the CGIAR Challenge Program on W... more This report was prepared as a visioning and discussion piece for the CGIAR Challenge Program on Water and Food's Ganges Basin Development Challenge. The views expressed in this report are those of the consultant and in no way are representative of the thinking of the Challenge Program on Water and Food.
Pakistan Journal of Biological Sciences, Jun 1, 2000
One month old fry of a planktivorous Indian major carp species, catla, Catla catla (Hamilton) was... more One month old fry of a planktivorous Indian major carp species, catla, Catla catla (Hamilton) was reared upto fingerling stage in earthen ponds in order to determine the food preference and selection using Ivlev's Electivity Index under two different treatments. Except Chlorophyceae, it showed positive selection to Cyanophyceae, Euglenophyceae, Rodophyceae, Crustacea and Rotifera. However, Catla, during its fry-fingerling stage exerted strong preference for zooplankton and almost neutral response to phytoplankton in both conditions.
Pakistan Journal of Biological Sciences
Tropical Pest Management, 1986
Experiments were conducted to find out suitable pesticides, dose and method of application to con... more Experiments were conducted to find out suitable pesticides, dose and method of application to control white grubs (Scarabid beetles) damaging sugarcane in northern and northwestern districts of Bangladesh. Pesticides tested were Dieldrin (50WP, 40WP, 20EC), Heptachlor (40WP, 20G), Aldrin 5D, Ekalux 5G (Quinalphos) and Furadan 3G (Carbofuran) at varying doses and methods. Furadan 3G at 2.0 kg a.i./ha and 1.5
"Le Design des interfaces numériques est un ouvrage co-écrit par l'APCI... more "Le Design des interfaces numériques est un ouvrage co-écrit par l'APCI, *Designers Interactifs*, Mov'eo et Systematic Paris-Region. Les interfaces numériques sont au cœur du déploiement de dispositifs interactifs, dans différents domaines comme les transports, les communications, les services... Pour travailler à la conception de nouveaux outils et devices (interfaces Homme-Machine, applications logicielles, sites Web, objets connectés…), les professionnels de ces différents secteurs doivent apprendre à parler un seul et même langage. Cet ouvrage rassemble les définitions des 170 mots-clés du design des interfaces numériques, dans ses domaines fondamentaux que sont le design d’interaction, l'ingénierie, l’ergonomie, l’architecture de l’information. Composé par autant de termes du quotidien que de concepts spécialisés, cet ouvrage s'inscrit dans une démarche pédagogique, à destination des professionnels (designers, ingénieurs, ergonomes…) et des étudiants."