AutoMTL: A Programming Framework for Automated Multi-Task Learning

• URL: http://arxiv.org/abs/2110.13076v1
• Date: Mon, 25 Oct 2021 16:13:39 GMT
• Title: AutoMTL: A Programming Framework for Automated Multi-Task Learning
• Authors: Lijun Zhang, Xiao Liu, Hui Guan
• Abstract summary: Multi-task learning (MTL) jointly learns a set of tasks. A major barrier preventing the widespread adoption of MTL is the lack of systematic support for developing compact multi-task models. We develop the first programming framework AutoMTL that automates MTL model development.
• Score: 23.368860215515323
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-task learning (MTL) jointly learns a set of tasks. It is a promising approach to reduce the training and inference time and storage costs while improving prediction accuracy and generalization performance for many computer vision tasks. However, a major barrier preventing the widespread adoption of MTL is the lack of systematic support for developing compact multi-task models given a set of tasks. In this paper, we aim to remove the barrier by developing the first programming framework AutoMTL that automates MTL model development. AutoMTL takes as inputs an arbitrary backbone convolutional neural network and a set of tasks to learn, then automatically produce a multi-task model that achieves high accuracy and has small memory footprint simultaneously. As a programming framework, AutoMTL could facilitate the development of MTL-enabled computer vision applications and even further improve task performance. Code of AutoMTL will be available at https://github.com/zhanglijun95/AutoMTL

Related papers

• DeeR-VLA: Dynamic Inference of Multimodal Large Language Models for Efficient Robot Execution [114.61347672265076]
  Development of MLLMs for real-world robots is challenging due to the typically limited computation and memory capacities available on robotic platforms. We propose a Dynamic Early-Exit Framework for Robotic Vision-Language-Action Model (DeeR) that automatically adjusts the size of the activated MLLM. DeeR demonstrates significant reductions in computational costs of LLM by 5.2-6.5x and GPU memory of LLM by 2-6x without compromising performance.
  arXiv  Detail & Related papers  (2024-11-04T18:26:08Z)
• AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML [56.565200973244146]
  Automated machine learning (AutoML) accelerates AI development by automating tasks in the development pipeline. Recent works have started exploiting large language models (LLM) to lessen such burden. This paper proposes AutoML-Agent, a novel multi-agent framework tailored for full-pipeline AutoML.
  arXiv  Detail & Related papers  (2024-10-03T20:01:09Z)
• AdapMTL: Adaptive Pruning Framework for Multitask Learning Model [5.643658120200373]
  AdapMTL is an adaptive pruning framework for multitask models. It balances sparsity allocation and accuracy performance across multiple tasks. It showcases superior performance compared to state-of-the-art pruning methods.
  arXiv  Detail & Related papers  (2024-08-07T17:19:15Z)
• AutoMMLab: Automatically Generating Deployable Models from Language Instructions for Computer Vision Tasks [39.71649832548044]
  AutoMMLab is a general-purpose LLM-empowered AutoML system that follows user's language instructions. The proposed AutoMMLab system effectively employs LLMs as the bridge to connect AutoML and OpenMMLab community. Experiments show that our AutoMMLab system is versatile and covers a wide range of mainstream tasks.
  arXiv  Detail & Related papers  (2024-02-23T14:38:19Z)
• AutoML-GPT: Automatic Machine Learning with GPT [74.30699827690596]
  We propose developing task-oriented prompts and automatically utilizing large language models (LLMs) to automate the training pipeline. We present the AutoML-GPT, which employs GPT as the bridge to diverse AI models and dynamically trains models with optimized hyper parameters. This approach achieves remarkable results in computer vision, natural language processing, and other challenging areas.
  arXiv  Detail & Related papers  (2023-05-04T02:09:43Z)
• AdaMTL: Adaptive Input-dependent Inference for Efficient Multi-Task Learning [1.4963011898406864]
  We introduce AdaMTL, an adaptive framework that learns task-aware inference policies for multi-task learning models. AdaMTL reduces the computational complexity by 43% while improving the accuracy by 1.32% compared to single-task models. When deployed on Vuzix M4000 smart glasses, AdaMTL reduces the inference latency and the energy consumption by up to 21.8% and 37.5%, respectively.
  arXiv  Detail & Related papers  (2023-04-17T20:17:44Z)
• ART: Automatic multi-step reasoning and tool-use for large language models [105.57550426609396]
  Large language models (LLMs) can perform complex reasoning in few- and zero-shot settings. Each reasoning step can rely on external tools to support computation beyond the core LLM capabilities. We introduce Automatic Reasoning and Tool-use (ART), a framework that uses frozen LLMs to automatically generate intermediate reasoning steps as a program.
  arXiv  Detail & Related papers  (2023-03-16T01:04:45Z)
• M$^3$ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design [95.41238363769892]
  Multi-task learning (MTL) encapsulates multiple learned tasks in a single model and often lets those tasks learn better jointly. Current MTL regimes have to activate nearly the entire model even to just execute a single task. We present a model-accelerator co-design framework to enable efficient on-device MTL.
  arXiv  Detail & Related papers  (2022-10-26T15:40:24Z)
• Controllable Pareto Multi-Task Learning [55.945680594691076]
  A multi-task learning system aims at solving multiple related tasks at the same time. With a fixed model capacity, the tasks would be conflicted with each other, and the system usually has to make a trade-off among learning all of them together. This work proposes a novel controllable multi-task learning framework, to enable the system to make real-time trade-off control among different tasks with a single model.
  arXiv  Detail & Related papers  (2020-10-13T11:53:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.