SDN optimizes flow entries to reduce control plane overhead.
problem Efficiently manage flow entries in SDN switches with limited TCAM capacity.
method Proposes reinforcement learning algorithms to optimize flow entries.
result Achieves up to 60% reduction in long-term control plane overhead.
SRG improves optimization efficiency with reduced memory and computation overhead.
problem Limited applicability of variance-reduced optimization algorithms.
method SRG (stochastic reweighted gradient) using importance sampling.
result SRG outperforms SGD and provably improves convergence in strongly-convex cases.
Study of 2D Lorentzian anti-de Sitter plane using geometric control theory.
problem Understanding extremal trajectories and reachable set on anti-de Sitter plane.
method Geometric control theory and differential geometry.
result Construction of optimal synthesis and description of Lorentzian distance.
EControl improves fast distributed optimization with compression and error control.
problem Stable convergence issues in distributed training with compression.
method Proposes EControl to regulate error compensation and prove fast convergence.
result Proves fast convergence for EControl in various convex settings without additional assumptions.
Automatically finds strong neural network topologies for continuous control tasks.
problem Handcrafted neural network architectures limit the performance of Deep Reinforcement Learning.
method Combines Neuroevolution with off-policy training and proposes a novel architecture mutation operator.
result The proposed Actor-Critic Neuroevolution algorithm often outperforms strong baseline methods.
New geometric approach controls motion of a spinning sphere on a plane.
problem Controlling the motion of a spinning sphere on a straight path.
method Geometric-based planning approach using Darboux frame kinematics.
result Iterative algorithm tunes inputs for desired configurations.
The paper solves optimal control problems for various convex sets using convex trigonometry.
problem Optimal control problems with 2D convex compact sets.
method Using convex trigonometry to derive extremals for various problems.
result Geodesics in multiple sub-Finsler problems are derived.
New kinematic model for a spin-rolling sphere using Darboux frame.
problem Control and planning of an underactuated sphere on a plane.
method Darboux frame transformation to fully-actuated model.
result Underactuated sphere model transformed into fully-actuated model.
A new method reduces communication in distributed RL without sacrificing performance.
problem High communication overhead in distributed RL systems.
method Adaptive policy gradient approach that skips communication during iterations.
result Reduces communication rounds needed for learning accuracy without degrading performance.
Classifies degenerations of complex projective plane with rational singularities.
problem Classifying singularities of complex projective plane.
method Assuming Wahl's conjecture, classifies degenerations using rational homology disk smoothing.
result Classifies surfaces with rational singularities, including new degenerations with non-log canonical singularities.
Differentiable cutting-plane layers solve parametric mixed-integer linear optimization problems.
problem Solving parametric mixed-integer linear optimization problems with changing data.
method Introducing cutting-plane layers (CPLs) for differentiable cutting-plane generation.
result The algorithm computes solutions with low integrality gaps and generalizes to unseen instances.
New boundary and point constraints for controlling conformal surfaces.
problem Controlling the geometry of surfaces defined by minimizers of conformal variational problems.
method Introducing new boundary conditions, point constraints, and flux constraints to control the metric and conformal scale factor.
result Introduces intuitive controls for exploring a subspace of conformal immersions.
CTRL improves reinforcement learning by combining control strategies.
problem Controlling agents to herd and contain target agents in a goal region.
method CTRL algorithm that enhances tabular learning algorithms with a tutoring model-based control strategy.
result Significantly reduces learning times and improves exploration of the state-space.
Agrachev's problem on circle turns is solved for various topologies.
problem How many times must a circle be turned to allow deformation with non-degenerate Frenet frame?
method Introduced decorated turn data to retain a nontrivial turn-counting problem. Analyzed different topologies and dimensions.
result For Cn curve topology, k(2)=1, k(3)=2, and k(n)=1 for n≥4. Spherical Fenchel obstruction in all dimensions n≥4. We propose a novel data-dependent structured gradient regularizer to increase the robustness of neural networks vis-a-vis adversarial perturbations. Our regularizer can be derived as a controlled approximation from first principles, leveraging the fundamental link between training with noise and regularization. It adds…
Optimizes deep learning training by treating it as an optimal control problem.
problem Fragility of deep neural networks to adversarial inputs.
method Formulates adversarial training as a min-max optimization problem and interprets it as an optimal control problem.
result Provides the first convergence analysis of adversarial training algorithm.
We show LLMs can be locally linear, enabling better control of activations.
problem Suboptimal control of LLM activations during generation.
method Model LLM inference as a linear dynamical system, compute feedback controllers using Jacobians, and adapt classical control theory.
result Robust, fine-grained control of LLM activations across models and tasks.
The hyperbolic structure of equilateral pentagons is mapped to a tiling of the hyperbolic plane.
problem Mapping the realization space of equilateral pentagons to a hyperbolic plane.
method Combining combinatorial correspondence, Riemann mapping theorem, and normalization procedure.
result A full conformal parameterization of the space of equilateral pentagons.
Develops diffusion samplers for target distributions with efficient score and density estimates.
problem Estimating scores and densities for time-varying distributions.
method Sequential Monte Carlo with diffusion paths and control variates.
result Effective samplers for time-varying distributions with theoretical guarantees and practical applications.
FedLog reduces communication in federated learning by sharing data summaries.
problem Significant communication overhead in federated learning with large model parameters.
method Shares minimal sufficient statistics via Bayesian inference and differential privacy.
result High learning accuracy with low communication overhead.
DriftLite improves inference quality of diffusion models without retraining.
problem Adapting pre-trained diffusion models to new target distributions without retraining.
method Lightweight, training-free particle-based approach that steers inference dynamics with optimal stability control.
result Consistently reduces variance and improves sample quality over existing methods.
New protocols implement logical gates on encoded qubits with minimal overhead.
problem Efficiently performing universal logical gates on encoded qubits with minimal overhead.
method Using topological codes associated to hyperbolic surfaces, we introduce protocols to implement Dehn twists through constant depth unitary circuits.
result Demonstrated the possibility of applying universal logical gate sets on encoded qubits through constant depth unitary circuits and with constant space overhead.
We present an algorithm for converting an indoor spherical panorama into a photograph with a simulated overhead view. The resulting image will have an extremely wide field of view covering up to 4π steradians of the spherical panorama. We argue that our method complements the stereographic projection commonly used in t…
Direct convolution eliminates memory overhead and improves performance.
problem Memory overhead and suboptimal performance in convolution layers.
method Implementing direct convolution without additional memory overhead.
result Performance improvement between 10% to 400% on various architectures.
Augmentations and sheaves linked for Legendrian graphs.
problem Understanding categorical Legendrian isotopy invariants.
method Equivalence between augmentation category and DG category of sheaves.
result Proved 'augmentations are sheaves' for Legendrian graphs.
A new method speeds up spectral normalization for neural nets.
problem Efficiently controlling the spectral norm of convolutional layers.
method Depthwise separable convolutions with spectral normalization.
result Significant reduction in computational and memory costs.
Deep learning reduces training overhead in massive MIMO systems.
problem Reducing training overhead in massive MIMO systems.
method Use of deep learning (NNs) to improve CSI acquisition and feedback processes.
result Significant improvements in performance and reduced complexity.
We consider the problem of learning a binary classifier from n different data sources, among which at most an η fraction are adversarial. The overhead is defined as the ratio between the sample complexity of learning in this setting and that of learning the same hypothesis class on a single data distribution. We pr…
An adaptive time-stepping controller improves stability and accuracy of ResNets.
problem Improving stability and performance of ResNets using adaptive time stepping.
method Developed an adaptive time-stepping controller based on Runge-Kutta-Fehlberg method.
result Demonstrated improved stability and accuracy of ResNets without additional overhead.
An optimal control problem associated with the dynamics of the orientation of a bipolar molecule in the plane can be understood by means of tools in differential geometry. For first time in the literature k-symplectic formalism is used to provide the optimal control problems associated to some families of partial dif…
Efficient communication improves MARL performance in StarCraft II.
problem Achieving efficient communication among agents in multi-agent reinforcement learning.
method Variance Based Control (VBC) to limit message variance and preserve useful information.
result VBC reduces communication overhead by 2-10 times compared to state-of-the-art MARL algorithms.
We study the collaborative PAC learning problem recently proposed in Blum et al.~\cite{BHPQ17}, in which we have k players and they want to learn a target function collaboratively, such that the learned function approximates the target function well on all players' distributions simultaneously. The quality of the col…
We prove the nonexistence of stable immersed minimal surfaces uniformly conformally equivalent to the complex plane in any complete orientable four-dimensional Riemannian manifold with uniformly positive isotropic curvature. We also generalize the same nonexistence result to higher dimensions provided that the ambient …
In this paper we investigate Uludag's method for constructing new curves whose fundamental groups are central extensions of the fundamental group of the original curve by finite cyclic groups. In the first part, we give some generalizations to his method in order to get new families of curves with controlled fundamenta…
New method reduces parameter overhead for Bayesian neural networks.
problem High parameter overhead and difficulty of implementation in variational Bayesian neural networks.
method Constructs a general variational family for ensemble-based Bayesian neural networks that works well with batch normalization layers.
result Improves predictive accuracy and achieves almost perfect calibration on a ResNet-18 trained with ImageNet.
In this paper we solve the following problem in the affirmative: Let Z be a continuum in the plane $\complex$ and suppose that $h:Z\times [0,1]\to\complex$ is an isotopy starting at the identity. Can h be extended to an isotopy of the plane? We will provide a new characterization of an accessible point in a planar …
A major contributing factor to the recent advances in deep neural networks is structural units that let sensory information and gradients to propagate easily. Gating is one such structure that acts as a flow control. Gates are employed in many recent state-of-the-art recurrent models such as LSTM and GRU, and feedforwa…
Configuration spaces of distinct labeled points on the plane are of practical relevance in designing safe control schemes for Automated Guided Vehicles (robots) in industrial settings. In this announcement, we consider the problem of the construction and classification of configuration spaces for graphs. Topological da…
This paper improves federated learning efficiency by adaptively sparsifying gradients.
problem Efficiently training machine learning models with geographically dispersed data.
method Adaptive gradient sparsification for non-i.i.d. local datasets, fairness-aware, online learning approach.
result Up to 40% improvement in model accuracy for a finite training time.
A novel controller for wheeled robots handles joystick inputs for smooth steering.
problem Steering control for differential-drive wheeled robots from indirect joystick inputs.
method Developed a geometric controller based on Darboux frame kinematics.
result Smooth trajectories achieved with safety constraints and no desired states.
New method builds hyperbolic spheres with controlled holonomy.
problem Creating hyperbolic spheres with specific holonomy properties.
method Gluing simple building blocks to form hyperbolic cone spheres.
result Any Deroin-Tholozan representation can be realized as cone sphere holonomy.
We use controlled topology applied to the action of the infinite dihedral group on a partially compactified plane and deduce two consequences for algebraic K-theory. The first is that the family in the K-theoretic Farrell-Jones conjecture can be reduced to only those virtually cyclic groups which admit a surjection wit…
Federated LIDAR aided beam selection reduces mmWave beam search overhead.
problem Efficient link configuration in mmWave communication systems with reduced beam search overhead.
method Federated learning of LIDAR data to train a shared neural network for beam selection.
result Proposed method significantly outperforms previous works in performance and complexity.
We study the following question: given a set P of 3d-2 points and an immersed curve G in the real plane R^2, all in general position, how many real rational plane curves of degree d pass through these points and are tangent to this curve. We count each such curve with a certain sign, and present an explicit formula for…
DeepMPC uses neural networks to control complex fluid flows efficiently.
problem Controlling complex fluid flows in real-time is challenging due to high dimensionality and multi-scale dynamics.
method Deep learning, specifically recurrent neural networks (RNNs), embedded in model predictive control (MPC) framework.
result Significant improvements in control performance achieved through online updates to prediction accuracy.
Classifying streaming data requires the development of methods which are computationally efficient and able to cope with changes in the underlying distribution of the stream, a phenomenon known in the literature as concept drift. We propose a new method for detecting concept drift which uses an Exponentially Weighted M…
Paper proposes SCALLION and SCAFCOM for compressed FL with reduced communication.
problem Reducing communication overhead in Federated Learning with data heterogeneity and partial participation.
method Revisit and simplify stochastic controlled averaging, proposing SCALLION and SCAFCOM for unbiased and biased compression.
result SCALLION and SCAFCOM outperform existing methods in communication and computation complexities.
EKF improves DNN uncertainty quantification with less computation.
problem Uncertainty quantification in deep neural networks.
method Extended Kalman Filtering applied to DNNs.
result EKF reduces computational overhead for uncertainty quantification.