AI summaryⓘ
The authors developed CAESAR-LDAR, a method to compress scientific data that involves multiple related measurements. Their approach improves compression by rearranging the compressed data to reduce dependencies and by using a predictive model that looks at nearby values to better understand remaining patterns. Tests show these two techniques help in different situations: one is better when there are strong linear relationships, and the other works well for local spatial patterns. Combining both methods gives the best compression results without much extra computation for one part, though the predictive model is slower. Overall, the authors suggest combining global and local strategies to better compress complex scientific datasets.
scientific compressionmultivariate datalatent spaceorthogonal transformautoregressive modelrate-distortionmatrix-exponential parameterizationcausal priorreconstruction tolerancecross-channel dependence
Authors
Liangji Zhu, Anand Rangarajan, Sanjay Ranka
Abstract
Scientific simulations generate collections of physical fields with heterogeneous statistics and dependencies, yet learned compressors often encode those fields independently or rely on a shared encoder without explicitly modeling the structure that remains in latent space. We present CAESAR-LDAR, an error-controlled multivariate learned compressor that augments a shared CAESAR-V backbone with two complementary mechanisms: a trainable orthogonal transform that reorganizes dependence across aligned latent channels, and a causal autoregressive hierarchical prior that captures local spatial structure left after transformation. Orthogonality is maintained through a matrix-exponential parameterization, making the transform exactly invertible without an additional penalty. A common residual-correction stage is applied uniformly to all variants to enforce the requested reconstruction tolerance. Experiments across combustion, climate, and turbulence data show that the two mechanisms are useful in different regimes. Latent decorrelation helps most when substantial linear cross-channel dependence survives the nonlinear encoder, whereas autoregressive modeling remains effective when the remaining structure is primarily local or spatial. Their combination provides the strongest or near-strongest rate-distortion performance across the evaluated datasets. The global transform adds little computational overhead, while autoregressive coding introduces a larger throughput tradeoff. More broadly, the results suggest a practical design principle for multivariate scientific compression: exploit global cross-channel dependence when it is measurably present in latent space, and use local probabilistic context as a complementary mechanism across a wider range of data regimes.