AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Sequential Attend, Infer, Repeat: Generative Modelling of Moving Objects

Adam Kosiorek, Hyunjik Kim, Yee Whye Teh, Ingmar Posner

NeurIPS 2018 · Probabilistic and scientific modeling › Molecules and proteins · approach: Neuro-symbolic probabilistic layers · 89 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add temporal propagation and discovery to AIRMLP-AIR · Moving multi-MNIST testlog marginal likelihood: 6443.6 -> 6617.6Improved
Add temporal propagation and discovery to AIRMLP-AIR · Moving multi-MNIST testcounting accuracy: 0.9058 -> 0.9986Improved
Add temporal propagation and discovery to AIRMLP-AIR · Moving multi-MNIST testaddition accuracy: 0.8644 -> 0.9998Improved
Use convolutional image/glimpse encoders and subpixel-convolution decoder instead of fully connected networksSQAIR · Moving multi-MNIST testlog marginal likelihood: 6617.6 -> 6784.8Improved
Use convolutional image/glimpse encoders and subpixel-convolution decoder instead of fully connected networksSQAIR · Moving multi-MNIST testaddition accuracy: 0.9998 -> 0.9990Mixed
Remove proposal-glimpse extraction in propagation (ablation)SQAIR · Moving multi-MNISTidentity consistencyHurt
Use raw observation access in propagation instead of an encoded proposal glimpse (ablation)SQAIR · Moving multi-MNISTidentity consistencyHurt
Use NVIL instead of VIMCO for discrete-latent gradient estimationSQAIR · Moving multi-MNISTtraining performanceHurt
Learn the output likelihood standard deviation rather than fixing it (ablation)SQAIR · Moving multi-MNISTtraining successHurt
Use a smaller network with 32 units per layer (ablation)SQAIR · Moving multi-MNISTquantitative performanceHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy