What if we use ensemble technique not for training but for prediction?
@brandmasis Жыл бұрын
I think this can help artists, to protect their art style.
@NocturnalCachalot3 жыл бұрын
The topics on this channel are just perfect, was thinking about exactly that quite recently 😊 As in: What will happen if an ANN is hacked and what will be the result/impact on the industry and other institutions in the long run (I can imagine well what will happen in case of the police using a "contaminated" computer vision system)? I rarely see these topics covered on any cyber security related blogs etc. So, thanks as always for satisfying my curious mind 😂👍
@JasonMayes3 жыл бұрын
Thank you for the kind words and it's great to hear your feedback. Are their any other topics you have been thinking about that you would like to know more about? Always good to know what people want to hear more about in case I find a good case study for it
@NocturnalCachalot3 жыл бұрын
@@JasonMayes My most recent thoughts (because of an NLP project: we use TensorFlow there as well) are about good solutions for syntactical trees to be parsed more accurately, as in subordinate and dependent clauses. Usually the parser (despite the tagger working correctly) has some issues with discerning parent nodes from child nodes. I have a few ideas, but will see how it goes 😁 If I have any other thoughts on different issues, I'll definitely write them under the next video 🙏
@JasonMayes3 жыл бұрын
@@NocturnalCachalot thank you!
@ghastlyo92843 жыл бұрын
This demo is amazing! XD
@JasonMayes3 жыл бұрын
Right! It is good to look into these things in ML's younger years
@carrycooldude3 жыл бұрын
Really Love the project, Hope so I can contribute to this project