www.fe.up.pt\/~tavares<\/a><\/span>) <\/div>\n <\/div>\n <\/p>\n <\/p>\n
Abstract:<\/p>\n
The segmentation of biomedical images by computational methods is very challenging, and it is mostly undertaken using thresholding, deformable models built on statistical, geometrical or physical principles, and\/or machine learning based approaches. Examples of current applications of segmentation methods include the identification of skin lesions, lungs, heart, prostate, liver, blood vessels, brain, ear, and related structures, just to name a few. In this lecture, algorithms that we have developed to segment images acquired using different biomedical imaging modalities will be described and their use in different applications discussed.<\/p>\n
\n
[\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row column_structure=”1_2,1_2″ use_custom_gutter=”on” gutter_width=”1″ make_equal=”on” _builder_version=”4.0.11″ _i=”2″ _address=”0.2″][et_pb_column type=”1_2″ _builder_version=”4.0.6″ background_color=”#dddddd” animation_style=”zoom” _i=”0″ _address=”0.2.0″][et_pb_image src=”http:\/\/cyprusconferences.org\/caip2021\/wp-content\/uploads\/2021\/08\/joao.tif” _builder_version=”3.29.3″ _i=”0″ _address=”0.2.0.0″][\/et_pb_image][et_pb_image src=”http:\/\/cyprusconferences.org\/caip2021\/wp-content\/uploads\/2021\/09\/dovrolis.jpg” align=”center” _builder_version=”3.29.3″ min_height=”339px” _i=”1″ _address=”0.2.0.1″][\/et_pb_image][et_pb_text _builder_version=”4.0.11″ custom_padding=”10px||10px||false|false” _i=”2″ _address=”0.2.0.2″]<\/p>\n
Constantine Dovrolis<\/h2>\n
Georgia Institute of Technology\u00a0(Georgia Tech)<\/p>\n
[\/et_pb_text][\/et_pb_column][et_pb_column type=”1_2″ _builder_version=”4.0.11″ background_color=”#efefef” animation_style=”zoom” _i=”1″ _address=”0.2.1″][et_pb_text _builder_version=”4.0.11″ header_4_font=”|600|||||||” custom_padding=”20px|10px|10px|10px|false|false” _i=”0″ _address=”0.2.1.0″]<\/p>\n
If the brain is a very sparse network, why does deep learning use dense neural networks?<\/h4>\n
\n
[\/et_pb_text][et_pb_text _builder_version=”4.0.11″ custom_padding=”|10px|10px|10px|false|false” _i=”1″ _address=”0.2.1.1″]<\/p>\n
Bio:\u00a0<\/p>\n
Dr. Constantine Dovrolis is a Professor at the School of Computer Science at the Georgia Institute of Technology (Georgia Tech). He is a graduate of the Technical University of Crete (Engr.Dipl. 1995), University of Rochester (M.S. 1996), and University of Wisconsin-Madison (Ph.D. 2000). His research combines Network Science, Data Mining and Machine Learning with applications in climate science, biology, neuroscience, sociology and machine learning. More recently, his group has been focusing on neuro-inspired architectures for machine learning based on what is currently known about the structure of brain networks.<\/p>\n
Abstract:<\/p>\n
What we know from neuroscience (\u201cconnectomics\u201d) is that the brain is, overall, a very sparse network with relatively small locally dense clusters of neurons. These topological properties are crucial for the brain\u2019s ability to perform efficiently, robustly, and to process information in a hierarchically modular manner. On the other hand, the artificial neural networks we use today are very dense, or even fully connected, at least between successive layers. Additionally,
it is well known that deep neural networks are highly over-parameterized: pruning studies have shown that it is often possible to eliminate 90% of the connections (weights) without significant loss in performance. Pruning, however, is typically performed after the dense network has been trained, which only improves the run-time efficiency of the inference process. The previous points suggest that we need methods to design sparse neural networks, without any training, that can perform almost as well as the corresponding dense networks after training. This talk will first provide some background in the pruning literature, either after training or before training. Then, we will present a recently proposed (ICML 2021) method called PHEW (Paths with Higher Edge Weights) which creates sparse neural networks,
before training, and that can learn fast and generalize well. Additionally, PHEW does not require access to any data as it only depends on the initial weights and the topology of the given network architecture.<\/p>\n
\n
[\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section]<\/p>\n","protected":false},"excerpt":{"rendered":"
Keynote SpeakersJo\u00e3o Manuel R. S. Tavares Faculdade de Engenharia da Universidade do Porto, PortugalBiomedical Imaging Segmentation: from thresholding to deep learning based methods Bio:\u00a0 Jo\u00e3o Manuel R. S. Tavares graduated in Mechanical Engineering at the Universidade do Porto, Portugal in 1992. He also earned his M.Sc. degree and Ph.D. degree in Electrical and Computer Engineering […]<\/p>\n","protected":false},"author":1,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_et_pb_use_builder":"on","_et_pb_old_content":"","_et_gb_content_width":""},"_links":{"self":[{"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/pages\/16"}],"collection":[{"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/comments?post=16"}],"version-history":[{"count":8,"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/pages\/16\/revisions"}],"predecessor-version":[{"id":1066,"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/pages\/16\/revisions\/1066"}],"wp:attachment":[{"href":"https:\/\/cyprusconferences.org\/caip2021\/wp-json\/wp\/v2\/media?parent=16"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}