Deep image priorDeep image prior is a type of convolutional neural network used to enhance a given image with no prior training data other than the image itself. A neural network is randomly initialized and used as prior to solve inverse problems such as noise reduction, super-resolution, and inpainting. Image statistics are captured by the structure of a convolutional image generator rather than by any previously learned capabilities.
Cot–caught mergerThe cot–caught merger, also known as the merger or low back merger, is a sound change present in some dialects of English where speakers do not distinguish the vowel phonemes in words like cot versus caught. Cot and caught (along with bot and bought, pond and pawned, etc.) is an example of a minimal pair that is lost as a result of this sound change. The phonemes involved in the cot–caught merger, the low back vowels, are typically represented in the International Phonetic Alphabet as /ɒ/ and /ɔ/, respectively (or, in North America, co-occurring with the father–bother merger, as /ɑ/ and /ɔ/).
