scientific-multi-task-learning

Community

Multi-task learning with shared representations.

Authornahisaho
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Efficiently train and generalize across multiple related prediction tasks by learning a shared representation and task-specific heads, reducing data requirements and improving sample efficiency.

Core Features & Use Cases

  • Hard/Soft Parameter Sharing, GradNorm gradient normalization, PCGrad gradient projection
  • Task balancing and auxiliary task design for robust multi-task training
  • OpenML integration for diverse multitask datasets

Quick Start

Train a shared encoder with task-specific heads to jointly optimize multiple tasks.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: scientific-multi-task-learning
Download link: https://github.com/nahisaho/satori/archive/main.zip#scientific-multi-task-learning

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.