# CNT (Conditioning on Noisy Targets): A new Algorithm for Leveraging Top-Down Feedback

arXiv: https://arxiv.org/abs/2210.09505 · HTML: https://mlsanity.com/paper/2210.09505/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 30 of its 43 references that are in the map.
- First posted on arXiv 18 Oct 2022.

## Builds on (most cited first, 30 of 30)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [Analyzing and Improving the Image Quality of StyleGAN](https://mlsanity.com/paper/1912.04958/) (2019)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains](https://mlsanity.com/paper/2006.10739/) (2020)
- [Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization](https://mlsanity.com/paper/1703.06868/) (2017)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Progressive Neural Networks](https://mlsanity.com/paper/1606.04671/) (2016)
- [Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering](https://mlsanity.com/paper/1707.07998/) (2017)
- [Cascaded Diffusion Models for High Fidelity Image Generation](https://mlsanity.com/paper/2106.15282/) (2021)
- [Deep Variational Information Bottleneck](https://mlsanity.com/paper/1612.00410/) (2016)
- [A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets](https://mlsanity.com/paper/1707.08819/) (2017)
- [A simple neural network module for relational reasoning](https://mlsanity.com/paper/1706.01427/) (2017)
- [Caffe: Convolutional Architecture for Fast Feature Embedding](https://mlsanity.com/paper/1408.5093/) (2014)
- [On The Power of Curriculum Learning in Training Deep Networks](https://mlsanity.com/paper/1904.03626/) (2019)
- [Noisy Networks for Exploration](https://mlsanity.com/paper/1706.10295/) (2017)
- [Curriculum Learning: A Survey](https://mlsanity.com/paper/2101.10382/) (2021)
- [Modulating early visual processing by language](https://mlsanity.com/paper/1707.00683/) (2017)
- [Human Pose Estimation with Iterative Error Feedback](https://mlsanity.com/paper/1507.06550/) (2015)
- [Generalized Denoising Auto-Encoders as Generative Models](https://mlsanity.com/paper/1305.6663/) (2013)
- [Learning to Combine Top-Down and Bottom-Up Signals in Recurrent Neural Networks with Attention over Modules](https://mlsanity.com/paper/2006.16981/) (2020)
- [An Optimistic Perspective on Offline Reinforcement Learning](https://mlsanity.com/paper/1907.04543/) (2019)
- [Knowledge transfer in deep block-modular neural networks](https://mlsanity.com/paper/1908.08017/) (2019)
- [On the Transferability of Representations in Neural Networks Between Datasets and Tasks](https://mlsanity.com/paper/1811.12273/) (2018)
- [Neural Function Modules with Sparse Arguments: A Dynamic Approach to Integrating Information across Layers](https://mlsanity.com/paper/2010.08012/) (2020)

## Abstract (arXiv)

We propose a novel regularizer for supervised learning called Conditioning on Noisy Targets (CNT). This approach consists in conditioning the model on a noisy version of the target(s) (e.g., actions in imitation learning or labels in classification) at a random noise level (from small to large noise). At inference time, since we do not know the target, we run the network with only noise in place of the noisy target. CNT provides hints through the noisy label (with less noise, we can more easily infer the true target). This give two main benefits: 1) the top-down feedback allows the model to focus on simpler and more digestible sub-problems and 2) rather than learning to solve the task from scratch, the model will first learn to master easy examples (with less noise), while slowly progressing toward harder examples (with more noise).