@incollection{Borhani2022a,
abstract = {Convolutional Neural Networks are very similar to ordinary Neural Networks from the previous chapter: they are made up of neurons that have learnable weights and biases. Each neuron receives some inputs, performs a dot product and optionally follows it with a non-linearity. The whole network still expresses a single differentiable score function: from the raw image pixels on one end to class scores at the other. And they still have a loss function (e.g. SVM/Softmax) on the last (fully-connected) layer and all the tips/tricks we developed for learning regular Neural Networks still apply. So what does change? ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture. These then make the forward function more efficient to implement and vastly reduce the amount of parameters in the network.},
address = {Cham},
author = {Borhani, Reza and Borhani, Soheila and Katsaggelos, Aggelos K.},
booktitle = {Fundamentals of Machine Learning and Deep Learning in Medicine},
doi = {10.1007/978-3-031-19502-0_5},
pages = {89--110},
publisher = {Springer International Publishing},
title = {{Linear Classification}},
url = {https://link.springer.com/10.1007/978-3-031-19502-0_5},
year = {2022}
}

{"_id":"eSb7HkWetp5YyHQBi","bibbaseid":"borhani-borhani-katsaggelos-linearclassification-2022","author_short":["Borhani, R.","Borhani, S.","Katsaggelos, A. K."],"bibdata":{"bibtype":"incollection","type":"incollection","abstract":"Convolutional Neural Networks are very similar to ordinary Neural Networks from the previous chapter: they are made up of neurons that have learnable weights and biases. Each neuron receives some inputs, performs a dot product and optionally follows it with a non-linearity. The whole network still expresses a single differentiable score function: from the raw image pixels on one end to class scores at the other. And they still have a loss function (e.g. SVM/Softmax) on the last (fully-connected) layer and all the tips/tricks we developed for learning regular Neural Networks still apply. So what does change? ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture. These then make the forward function more efficient to implement and vastly reduce the amount of parameters in the network.","address":"Cham","author":[{"propositions":[],"lastnames":["Borhani"],"firstnames":["Reza"],"suffixes":[]},{"propositions":[],"lastnames":["Borhani"],"firstnames":["Soheila"],"suffixes":[]},{"propositions":[],"lastnames":["Katsaggelos"],"firstnames":["Aggelos","K."],"suffixes":[]}],"booktitle":"Fundamentals of Machine Learning and Deep Learning in Medicine","doi":"10.1007/978-3-031-19502-0_5","pages":"89–110","publisher":"Springer International Publishing","title":"Linear Classification","url":"https://link.springer.com/10.1007/978-3-031-19502-0_5","year":"2022","bibtex":"@incollection{Borhani2022a,\nabstract = {Convolutional Neural Networks are very similar to ordinary Neural Networks from the previous chapter: they are made up of neurons that have learnable weights and biases. Each neuron receives some inputs, performs a dot product and optionally follows it with a non-linearity. The whole network still expresses a single differentiable score function: from the raw image pixels on one end to class scores at the other. And they still have a loss function (e.g. SVM/Softmax) on the last (fully-connected) layer and all the tips/tricks we developed for learning regular Neural Networks still apply. So what does change? ConvNet architectures make the explicit assumption that the inputs are images, which allows us to encode certain properties into the architecture. These then make the forward function more efficient to implement and vastly reduce the amount of parameters in the network.},\naddress = {Cham},\nauthor = {Borhani, Reza and Borhani, Soheila and Katsaggelos, Aggelos K.},\nbooktitle = {Fundamentals of Machine Learning and Deep Learning in Medicine},\ndoi = {10.1007/978-3-031-19502-0_5},\npages = {89--110},\npublisher = {Springer International Publishing},\ntitle = {{Linear Classification}},\nurl = {https://link.springer.com/10.1007/978-3-031-19502-0_5},\nyear = {2022}\n}\n","author_short":["Borhani, R.","Borhani, S.","Katsaggelos, A. K."],"key":"Borhani2022a","id":"Borhani2022a","bibbaseid":"borhani-borhani-katsaggelos-linearclassification-2022","role":"author","urls":{"Paper":"https://link.springer.com/10.1007/978-3-031-19502-0_5"},"metadata":{"authorlinks":{}}},"bibtype":"incollection","biburl":"https://sites.northwestern.edu/ivpl/files/2023/06/IVPL_Updated_publications-1.bib","dataSources":["zFPgsTDAW8aDnb5iN","E6Bth2QB5BYjBMZE7","nbnEjsN7MJhurAK9x","PNQZj6FjzoxxJk4Yi","7FpDWDGJ4KgpDiGfB","bod9ms4MQJHuJgPpp","QR9t5P2cLdJuzhfzK","D8k2SxfC5dKNRFgro","7Dwzbxq93HWrJEhT6","qhF8zxmGcJfvtdeAg","fvDEHD49E2ZRwE3fb","H7crv8NWhZup4d4by","DHqokWsryttGh7pJE","vRJd4wNg9HpoZSMHD","sYxQ6pxFgA59JRhxi","w2WahSbYrbcCKBDsC","XasdXLL99y5rygCmq","3gkSihZQRfAD2KBo3","t5XMbyZbtPBo4wBGS","bEpHM2CtrwW2qE8FP","teJzFLHexaz5AQW5z"],"keywords":[],"search_terms":["linear","classification","borhani","borhani","katsaggelos"],"title":"Linear Classification","year":2022}