The embedding are a number given to a word that identifies the object. This number is given by a supervisor. Example the word man is given a 0.5, a woman is given a 0.51. A male monkey is given 0.3. A female monkey is 0.31. The vector of change from monkey to man is around 0.2. The distance of change can be plotted on a map, with t-sne.
PCA vs t-SNE: which one should you use for visualization: https://medium.com/analytics-vidhya/pca-vs-t-sne-17bcd882bf3d#:~:text=t%2DSNE%20is%20also%20a,large%20pairwise%20distance%20maximize%20variance.&text=It%20takes%20a%20set%20of,it%20into%20low%20dimensional%20data. Their work, by the others, is really new. And by the time you figure it out they will have published five more papers. Their intent is to keep you behind them in the race. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Tefaeb8e790a54cec-Mab6fd60c6ae453e823893963 Delivery options: https://agi.topicbox.com/groups/agi/subscription
