{
  "id": "2301.01572",
  "version": "v1",
  "published": "2023-01-04T12:48:05.000Z",
  "updated": "2023-01-04T12:48:05.000Z",
  "title": "Multi-Task Learning with Prior Information",
  "authors": [
    "Mengyuan Zhang",
    "Kai Liu"
  ],
  "comment": "Accepted to SDM'23",
  "categories": [
    "cs.LG",
    "cs.AI"
  ],
  "abstract": "Multi-task learning aims to boost the generalization performance of multiple related tasks simultaneously by leveraging information contained in those tasks. In this paper, we propose a multi-task learning framework, where we utilize prior knowledge about the relations between features. We also impose a penalty on the coefficients changing for each specific feature to ensure related tasks have similar coefficients on common features shared among them. In addition, we capture a common set of features via group sparsity. The objective is formulated as a non-smooth convex optimization problem, which can be solved with various methods, including gradient descent method with fixed stepsize, iterative shrinkage-thresholding algorithm (ISTA) with back-tracking, and its variation -- fast iterative shrinkage-thresholding algorithm (FISTA). In light of the sub-linear convergence rate of the methods aforementioned, we propose an asymptotically linear convergent algorithm with theoretical guarantee. Empirical experiments on both regression and classification tasks with real-world datasets demonstrate that our proposed algorithms are capable of improving the generalization performance of multiple related tasks.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2023-01-04T12:48:05.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "multi-task learning",
      "prior information",
      "multiple related tasks",
      "non-smooth convex optimization problem",
      "generalization performance"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}