REVIEW 30 cited by
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Deep Sets
read the original abstract
We study the problem of designing models for machine learning tasks defined on \emph{sets}. In contrast to traditional approach of operating on fixed dimensional vectors, we consider objective functions defined on sets that are invariant to permutations. Such problems are widespread, ranging from estimation of population statistics \cite{poczos13aistats}, to anomaly detection in piezometer data of embankment dams \cite{Jung15Exploration}, to cosmology \cite{Ntampaka16Dynamical,Ravanbakhsh16ICML1}. Our main theorem characterizes the permutation invariant functions and provides a family of functions to which any permutation invariant objective function must belong. This family of functions has a special structure which enables us to design a deep network architecture that can operate on sets and which can be deployed on a variety of scenarios including both unsupervised and supervised learning tasks. We also derive the necessary and sufficient conditions for permutation equivariance in deep models. We demonstrate the applicability of our method on population statistic estimation, point cloud classification, set expansion, and outlier detection.
Forward citations
Cited by 30 Pith papers
-
Understanding Truncated Positional Encodings for Graph Neural Networks
Truncation causes different positional encoding families to have unequal expressive power in GNNs, with truncated spectral PEs limited to 1-WL strength, and mixing families improves results on real datasets.
-
Meson-Nucleus Bound States with Neural-Network Quantum States
Neural-network quantum states applied to HAL QCD meson-nucleon potentials predict bound states for phi at A>=2, J/psi at A>=4, and eta_c at A>=6, with binding energies from tens of MeV to sub-MeV scales.
-
Adversarial Water-Filling: Theory, Algorithms and Foundation Model
Adversarial Water-Filling formulates competitive spectrum sharing as a constrained minimax problem and introduces a permutation-invariant GNN foundation model that approximates its stationary solutions with local line...
-
Search for low-mass resonances decaying to $\tau\tau$ and measurement of the $\Upsilon$ $\to$ $\tau\tau$ decay in proton-proton collisions at $\sqrt{s}$ = 13.6 TeV
CMS measures Υ(1S,2S,3S) → τ⁺τ⁻ with 5.8σ significance and cross section 3.5 ± 0.7 (stat) ± 0.7 (syst) nb while setting 95% CL upper limits of 40-400 pb on spin-zero resonance production to ττ with no observed excess ...
-
Patch Hierarchical Attention Transformer for Efficient Particle Jet Tagging
PHAT-JeT combines geometric message-passing with hierarchical patch attention to reach state-of-the-art accuracy and background rejection among resource-constrained jet tagging models on four benchmarks.
-
Higgs Physics with the XFEL Compton $\boldsymbol{\gamma\gamma}$ Collider Concept at $\boldsymbol{\sqrt{s}=125}$ GeV
An XFEL Compton gamma-gamma collider at 125 GeV with a set transformer deep learning classifier on particle-flow point clouds can achieve high-precision Higgs measurements across hadronic, semi-leptonic, and leptonic ...
-
Modeling isotropic polyconvex hyperelasticity by neural networks -- sufficient and necessary criteria for compressible and incompressible materials
CSSV-NNs and inc-CSSV-NNs provide universal approximation of frame-indifferent isotropic polyconvex hyperelastic energies, showing Ball's criterion is sufficient but not necessary.
-
Fermi Sets: Universal and interpretable neural architectures for fermions
Fermi Sets achieve universal approximation of fermionic wavefunctions using K antisymmetric bases times symmetric neural networks, where K equals 1 in 1D, 2 in 2D, and grows linearly with particle number in higher dimensions.
-
Monotone and Separable Set Functions: Characterizations and Neural Models
Characterizes monotone separating set functions with dimension bounds, proves non-existence on infinite domains, and introduces a Holder-stable neural model with a weak version of the property for universal monotone a...
-
Heavy-Flavor Electron Classification Using Hadronic Environment as Point Cloud
Point-cloud ML models classify charm- and bottom-origin electrons at ~80% purity for 40% efficiency, outperforming a BDT baseline, with performance limited by intrinsic decay similarity.
-
Simulation-Based Inference for Cluster Cosmology with Set-Based Neural Network Architectures
SBI framework with GNN-on-sets and masked autoregressive flow recovers input cosmologies from eRASS1 mocks at 11.5% precision on Ω_m and 4.4% on σ_8 using 3259 clusters.
-
Velocityformer: Broken-Symmetry-Matched Equivariant Graph Transformers for Cosmological Velocity Reconstruction
Velocityformer achieves 35% higher velocity correlation than linear theory by matching graph transformer inductive bias to the line-of-sight broken symmetry and conditioning on long-wavelength physics, while training ...
-
It Just Takes Two: Scaling Amortized Inference to Large Sets
A mean-pool deep set trained on sets of size at most two produces an encoder that generalizes to arbitrary sizes, decoupling representation learning from posterior modeling and making training cost independent of depl...
-
Tokenised Flow Matching for Hierarchical Simulation Based Inference
TFMPE combines likelihood factorisation with tokenised flow matching to enable efficient hierarchical SBI from single-site simulations, producing well-calibrated posteriors at lower computational cost on a new benchma...
-
Temporally Extended Mixture-of-Experts Models
Temporally extended MoE layers using the option-critic framework with deliberation costs cut switching rates below 5% while retaining most capability on MATH, MMLU, and MMMLU.
-
Diffusion-Based Point-Cloud Generation of Heavy-Ion Events
A two-stage score-driven diffusion model with Point-Edge Transformer generates realistic high-multiplicity heavy-ion events as point clouds.
-
Signal Decomposition Reveals Structure in Insider Threat Detection under Sparse Temporal Data
Separating presence from magnitude in sparse temporal audit data lets a dual-channel autoencoder focus learning on anomalous activity for insider threat detection.
-
Two-Layer Reinforcement Learning-Assisted Joint Beamforming and Trajectory Optimization for Multi-UAV Downlink Communications
A hierarchical GNN-RL framework for joint beamforming and trajectory optimization in multi-UAV systems outperforms baselines in sum rate, convergence, and generalization.
-
Drawback of Enforcing Equivariance and its Compensation via the Lens of Expressive Power
Enforcing equivariance reduces expressive power in 2-layer ReLU networks but enlarging the model compensates with proven size bounds and yields lower hypothesis space dimensionality for better generalization.
-
Efficient Inference for Coupled Hidden Markov Models in Continuous Time and Discrete Space
Proposes Latent Interacting Particle Systems with an efficient parameterization of twist potentials to enable approximate posterior inference for coupled continuous-time hidden Markov models via twisted sequential Mon...
-
Thermodynamically consistent machine learning model for excess Gibbs energy
HANNA is a thermodynamically consistent ML model for predicting excess Gibbs energy from molecular structures, trained on various binary mixture data and extended to multi-component mixtures using geometric projection.
-
First evidence for mixing-induced $CP$ violation in B$^0_\mathrm{s}$ $\to$ J/$\psi\,\phi$(1020) decays in pp collisions at $\sqrt{s} = $ 13 TeV
First evidence for non-zero φ_s in B_s^0 → J/ψ φ decays at 3.2σ from CMS 13 TeV data combined with prior 8 TeV result.
-
Population-Aware Physics-Informed Neural Particle Flow for Bayesian Update
PA-PINPF adds Deep Sets population encoders (state or feature) to PINPF for better Bayesian posterior particle transport on range-measurement and TDOA tasks.
-
Spatially Aware Linear Transformer (SAL-T) for Particle Jet Tagging
SAL-T enhances the linformer with spatially aware kinematic partitioning and convolutions to match full-attention transformer performance on jet tagging while keeping linear complexity and lower latency.
-
Container Unloading via Reinforcement Learning: Picking Order, Deadlock Avoidance, and Proof-of-Concept Simulation
Masked deep Q-learning in a custom simulation achieves 60% average success at item selection for container unloading, outperforming a 20% random baseline.
-
Statistical Properties of Training & Generalization
Neural scaling laws in deep learning interact with physics constraints and inductive biases beyond classical statistics.
-
The LHCb Experiment
This review summarizes the historical motivation, detector design, experimental techniques, and major physics results of the LHCb experiment at the LHC.
-
A Review of Diffusion-based Simulation-Based Inference: Foundations and Applications in Non-Ideal Data Scenarios
A synthesis of diffusion-based simulation-based inference methods that address model misspecification, irregular observations, and missing data in scientific applications.
-
Software and computing for Run 3 of the ATLAS experiment at the LHC
ATLAS reports on its Run 3 software infrastructure for data management, workflows, databases, validation, and physics analysis tools at the LHC.
-
Statistical Properties of Training & Generalization
Review of neural scaling laws and their relation to constraints and inductive biases when applying machine learning to physics problems.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.