Problem 3. (2 pts) Finding a separating hyperplane One of the canonical problems in machine learning is the one of finding a separating hyperplane. The input is composed of m sample-points F .m E Rd (cach point is composed of d numeric features) the where cach point comes with a corresponding label li {+1,-1} (so some of the sample points are labeled as positive, and some as negative). Ideally, in such a setting, one would like to find a separating hyperplane -in our "language": a constraint and a scalar b such that byperplane -in orative). Ideally, in for each sample point lisign.i -b) ie., all points with h = +1 are such that tF-a i > b; and all points labeled as negative are such that u-ii b; and all points labeled as negative are such that u-ii