Subject
Computer Science
Topic
Artificial Intelligence
A convolutional network slides the same small filter across the whole image (weight sharing). What is the main advantage?
A convolutional network slides the same small filter across the whole image (weight sharing). What is the main advantage?
a)
A separate weight is learned for every pixelb)
Perfect accuracy is guaranteedc)
No training data is neededd)
Far fewer parameters, and the same feature can be detected anywhere in the imageExplanation
One filter is reused at every position, so the network needs far fewer weights, and a pattern such as an edge is found wherever it appears.Related questions
A system must find every person and draw a box around each one in a photograph. Which task is required?A medical application must mark the exact pixels belonging to a tumour. Which vision task fits?The word “banks” must be mapped to the dictionary form “bank.” Which process is intended?A chatbot must identify “Dhaka” as a location in a sentence. Which NLP component applies?The sentence “I saw the man with a telescope” has two possible structures. What problem is illustrated?A search engine must return documents with similar meaning despite different words. Which representation is useful?A meeting transcript must mark which participant spoke each segment. Which speech-language task is required?Why is data augmentation such as small rotations useful in image training?