Search for: All records

Molecular representation learning has emerged as a promising approach for modeling molecules with deep learning in chemistry and beyond. While 3D geometric models effectively capture molecular structure, they typically process single static conformers, overlooking the inherent flexibility and dynamics of molecules. In reality, many molecular properties depend on distributions of thermodynamically accessible conformations rather than single structures. Recent works show that learning from conformer ensembles can improve molecular representations, but existing approaches either produce unphysical structures through averaging or require restrictive molecular alignment. In this paper, we propose Symmetry-Preserving Conformer Ensemble networks (SPiCE), which introduces two key innovations: (1) geometric mixture-of-experts for selective processing of scalar and vector features, and (2) hierarchical ensemble encoding that combines ensemble level representation with cross-conformer integration. Crucially, SPiCE ensures physically meaningful representations by maintaining joint equivariance to geometric transformations of individual conformers and conformer permutations. Extensive experiments demonstrate that SPiCE consistently outperforms existing conformer ensemble methods and state-of-the-art structural aggregation models across quantum mechanical and biological property prediction tasks. 

High synchronization overhead in frameworks like GNU OpenMP impedes fine-grained task parallelism on many-core architectures. We introduce three advances to GNU OpenMP: a lock-less concur- rent queue (XQueue), a scalable distributed tree barrier, and two NUMA-aware, lock-less load balancing strategies. Evaluated with Barcelona OpenMP Task Suite (BOTS) bench- marks, our XQueue and tree barrier improve performance by up to 1522.8×over the original GNU OpenMP. The load balancing strate- gies provide an additional performance improvement of up to 4×. We further apply these techniques to the TaskFlow runtime, demon- strating performance and scalability gains in selected applications while also analyzing the inherent limitations of the lock-less ap- proach on x86 architectures. 

Achieving efficient task parallelism on many-core architectures is an important challenge. The widely used GNU OpenMP implementation of the popular OpenMP parallel program- ming model incurs high overhead for fine-grained, short-running tasks due to time spent on runtime synchronization. In this work, we introduce and analyze three key advances that collectively achieve significant performance gains. First, we introduce XQueue, a lock-less concurrentqueueimplementationtoreplaceGNU’sprioritytaskqueueandremovetheglobal task lock. Second, we develop a scalable, efficient, and hybrid lock-free/lock-less distributed tree barrier to address the high hardware synchronization overhead from GNU’s central- ized barrier. Third, we develop two lock-less and NUMA-aware load balancing strategies. We evaluate our implementation using Barcelona OpenMP Task Suite (BOTS) benchmarks. Results from the first and second advances demonstrate up to 1522.8×performance im- provement compared to the original GNU OpenMP. Further improvements from lock-less load balancing show up to 4×improvement compared to GNU OpenMP using XQueue. Through a rich set of profiling and instrumentation tools, we are able to investigate the runtime behavior of GNU OpenMP and improve its performance on fine-grained tasks by many orders of magnitude.