paper-with-me

홈 › Papers

Eliminating Catastrophic Interference with Biased Competition

2020-07-03 · Amelia Elizabeth Pollard, Jonathan L. Shapiro

We present here a model to take advantage of the multi-task nature of complex datasets by learning to separate tasks and subtasks in and end to end manner by biasing competitive interactions in the network. This method does not require additional labelling or reformatting of data in a dataset. We propose an alternate view to the monolithic one-task-fits-all learning of multi-task problems, and describe a model based on a theory of neuronal attention from neuroscience, proposed by Desimone. We create and exhibit a new toy dataset, based on the MNIST dataset, which we call MNIST-QA, for testing Visual Question Answering architectures in a low-dimensional environment while preserving the more difficult components of the Visual Question Answering task, and demonstrate the proposed network architecture on this new dataset, as well as on COCO-QA and DAQUAR-FULL. We then demonstrate that this model eliminates catastrophic interference between tasks on a newly created toy dataset and provides competitive results in the Visual Question Answering space. We provide further evidence that Visual Question Answering can be approached as a multi-task problem, and demonstrate that this new architecture based on the Biased Competition model is capable of learning to separate and learn the tasks in an end-to-end fashion without the need for task labels.

📄 PDF Abstract BibTeX arXiv:2007.02833

Code (0)

등록된 구현이 없습니다.

Tasks

Question AnsweringVisual Question AnsweringVisual Question Answering (VQA)

Similar Papers 제목 키워드 기반

Online DPO: Online Direct Preference Optimization with Fast-Slow Chasing

2024-06-08 · Biqing Qi, Pengfei Li, Fangyuan Li, Junqi Gao 외

Direct Preference Optimization (DPO) improves the alignment of large language models (LLMs) with human values by training directly on human preference datasets, eliminating the need for reward models. However, due to the…

Continual Learning

A good body is all you need: avoiding catastrophic interference via agent architecture search

2021-08-18 · Joshua Powers, Ryan Grindle, Lapo Frati, Josh Bongard

In robotics, catastrophic interference continues to restrain policy training across environments. Efforts to combat catastrophic interference to date focus on novel neural architectures or training methods, with a recent…

All

Overcoming Catastrophic Interference by Conceptors

2017-07-16 · Xu He, Herbert Jaeger

Catastrophic interference has been a major roadblock in the research of continual learning. Here we propose a variant of the back-propagation algorithm, "conceptor-aided back-prop" (CAB), in which gradients are shielded …

Continual Learning

Overcoming Catastrophic Interference using Conceptor-Aided Backpropagation

2018-01-01 · ICLR 2018 1 · Xu He, Herbert Jaeger

Catastrophic interference has been a major roadblock in the research of continual learning. Here we propose a variant of the back-propagation algorithm, "Conceptor-Aided Backprop" (CAB), in which gradients are shielded b…

Continual LearningPermuted-MNIST

Learning to Modulate Random Weights: Neuromodulation-inspired Neural Networks For Efficient Continual Learning

2022-04-08 · Jinyung Hong, Theodore P. Pavlic

Existing Continual Learning (CL) approaches have focused on addressing catastrophic forgetting by leveraging regularization methods, replay buffers, and task-specific components. However, realistic CL solutions must be s…

Computational EfficiencyContinual LearningMeta-Learning