I'd like to optimize a 1x100 vector with Stochastic Gradient or Adam or something similar (for non deep-learning purposes). Can anyone recommend a relatively lightweight Clojure (or Java) way to do it? (I see there are https://mxnet.apache.org/versions/1.9.1/api/python/docs/tutorials/packages/optimizer/index.html#What-should-I-use?).