3 ms·

I'd be interested in knowing how much deep learning is changing the algorithms used in this field, given the performance of restricted boltzmann machines on the

• colincsl 13y ago
  I am surprised that there is no mention of this in the course syllabus -- in fact it looks like a lot of recent techniques that are missing. They don't even talking about LSA(/SVD)-based methods until the end of the course.
  • jere 13y ago
    Dumb, general question: should it really be a surprise that an Introduction course doesn't use all the recent techniques?
    • colincsl 13y ago
      I guess I often assume that topics classes like these assume some sort of background in machine learning. Perhaps that's not the case.
    • textminer 13y ago
      LSA and truncated SVDs on a user-item matrix date back to the 1980s, don't they?
• elq 13y ago
  You'd be surprised how overwhelmingly common and effective very simple methods like logistic regression and basic decision trees are for such systems. Further, RBMs and other deep learning tools require a significantly more sophisticated mathematical background than algebra and a much broader understanding overall.
• Irishsteve 13y ago
  The netflix prize touched on one of many areas related to recommender systems. As mentioned already, very simple methods can be really effective. Things such as the UI are also known to have a big impact on how 'useful' people find the recs.
• msellout 13y ago
  Not much yet. It's not well enough understood yet.