# Train a denoiser and sample iteratively Canonical: https://semiaifoundry.com/learn/atlas/denoising-generative-objectives/ Revision: 2026-10-07 Rights: https://semiaifoundry.com/ai-atlas/licenses.html model-lifecycle-multimodal · Atlas v1.3.0 # Train a denoiser and sample iteratively Compute a noisy training target and distinguish noise regression from a flow vector field. [Explore in Atlas](https://semiaifoundry.com/ai-atlas/#denoising-generative-objectives)Mathematical, data and behavior space; logical dimensions are not physical distance Learning objectiveCompute a noisy training target and distinguish noise regression from a flow vector field. ## The mechanism Generative training can operate on continuous values rather than discrete next-token labels. In a DDPM noise-prediction objective, a clean example is mixed with Gaussian noise at a chosen level and the model predicts that noise. Generation later runs a specified reverse sampling process; a single good denoising prediction is not a complete generated sample. Flow matching instead regresses a vector field along a declared probability path. The target, time convention and sampler therefore differ even when both start from noise. Neither mechanism is implemented by the tiny causal decoder. 1. Specify whether time indexes a discrete diffusion level or a continuous path. 2. Construct the noisy/interpolated input and exact regression target from retained random values. 3. Compute the declared coordinate-sum or coordinate-mean loss. 4. Evaluate samples from a complete stated sampler with fixed seeds and quality metrics separate from regression loss. ## Science and assumptions ### Train a denoiser and sample iteratively · worked relation $$ x_t=\sqrt{\bar\alpha_t}x_0+\sqrt{1-\bar\alpha_t}\epsilon,\quad L_\epsilon=\lVert\epsilon-\epsilon_\theta(x_t,t)\rVert^2,\qquad x_t^{flow}=(1-t)x_0+t x_1,\quad u=x_1-x_0 $$ Compute a noisy training target and distinguish noise regression from a flow vector field. Symbols and units x\_0 Clean diffusion value or explicitly named flow start dimensionless x\_1 Data endpoint in the flow example dimensionless alpha bar\_t Cumulative diffusion signal factor dimensionless epsilon Sampled standard Gaussian noise dimensionless t Declared diffusion index or normalized flow time index or dimensionless u Straight-path target velocity value per normalized time L\_epsilon Coordinate-summed noise error squared value For DDPM-style training, x0=2, cumulative alpha=0.64 and sampled noise epsilon=−1 give xt=0.8×2+0.6×(−1)=1. A noise prediction −0.5 has squared error 0.25. In a separate straight flow path from noise x0=−1 to data x1=2, time t=0.25 gives xt=−0.25 and target velocity x1−x0=3. The noise target −1 and flow target 3 refer to different constructions. Assumptions One-dimensional authored samples; diffusion and flow use separate time/start conventions explicitly. No complete sampler, likelihood bound or image-quality benchmark is executed. ## Engineering example For DDPM-style training, x0=2, cumulative alpha=0.64 and sampled noise epsilon=−1 give xt=0.8×2+0.6×(−1)=1. A noise prediction −0.5 has squared error 0.25. In a separate straight flow path from noise x0=−1 to data x1=2, time t=0.25 gives xt=−0.25 and target velocity x1−x0=3. The noise target −1 and flow target 3 refer to different constructions. ## Distinctions and tradeoffs It is a mistake to conclude: All noise-based objectives must have the same target. Continuous objectives offer different representation and sampling choices. More sampling steps incur additional work but do not alone guarantee useful or faithful outputs. Teach modality-specific representation, objective and evaluation choices. Compute a noisy training target and distinguish noise regression from a flow vector field. ## Specifications and interfaces Learning mode worked example The executable decoder has 423 parameters; worked extensions do not imply an additional simulator. Worked contract Compute a noisy training target and distinguish noise regression from a flow vector field. One-dimensional authored samples; diffusion and flow use separate time/start conventions explicitly. No complete sampler, likelihood bound or image-quality benchmark is executed. ### Interfaces Inputs and state Input Specify whether time indexes a discrete diffusion level or a continuous path. Result and check Output Compute a noisy training target and distinguish noise regression from a flow vector field. ## Check your understanding ### Why are the two regression targets different? 1. The flow velocity is the model’s next-token probability. 2. They supervise different declared paths and predicted quantities. 3. All noise-based objectives must have the same target. Answer and reasoning They supervise different declared paths and predicted quantities. They supervise different declared paths and predicted quantities. One-dimensional authored samples; diffusion and flow use separate time/start conventions explicitly. No complete sampler, likelihood bound or image-quality benchmark is executed. ## Evidence and scope Mode: worked example. One-dimensional authored samples; diffusion and flow use separate time/start conventions explicitly. No complete sampler, likelihood bound or image-quality benchmark is executed. Numerical examples are authored calculations, except explicitly identified automated core fixtures. Primary sources support the scoped mechanism; no human specialist review or learner outcome is implied. Model lifecycle studio: worked example. One-dimensional authored samples; diffusion and flow use separate time/start conventions explicitly. No complete sampler, likelihood bound or image-quality benchmark is executed. ## Sources - [Denoising Diffusion Probabilistic Models](https://arxiv.org/html/2006.11239v2) - [Flow Matching for Generative Modeling](https://arxiv.org/html/2210.02747v2) [Published lesson data](https://semiaifoundry.com/ai-atlas/lesson-data/denoising-generative-objectives-d5f093d00f1eca4e.json) · [Atlas releases and licensing](https://semiaifoundry.com/ai-atlas/licenses.html) ## Continue exploring ### Foundations - [Loss, gradients & optimizer](https://semiaifoundry.com/learn/atlas/loss-optimizer/) - [Evaluation & the AGI frontier](https://semiaifoundry.com/learn/atlas/evaluation/) ### Related lessons - [Pair images and captions in a shared representation](https://semiaifoundry.com/learn/atlas/image-text-contrastive-learning/) - [Patches, projection and visual instruction tuning](https://semiaifoundry.com/learn/atlas/visual-language-bridge/) - [Audio representation, alignment and transcript loss](https://semiaifoundry.com/learn/atlas/speech-audio-sequence-objectives/) - [Predict temporal representations without generating pixels](https://semiaifoundry.com/learn/atlas/video-latent-prediction/)