Improved batched SH algorithm maintains original performance.
problem Maintaining performance in batched multi-armed bandits.
method Simple batch version of Sequential Halving algorithm.
result Batching does not degrade SH algorithm's performance.
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,982 papers · 148 categories
Trend · papers per month
48 results for “halving”
Bitcoin's price indicators fail to predict halving epochs, but its time structure remains consistent.
problem The failure of widely used Bitcoin price indicators to predict halving epochs.
method Retrospective analysis of Bitcoin price indicators over four halving epochs, using a fixed mechanical rule.
result Price indicators fail to predict halving epochs, but Bitcoin's time structure remains consistent.
New algorithms optimize multiple machine learning metrics in real-world tasks.
problem Optimizing multiple conflicting performance criteria in real-world applications.
method Extends ASHA to multi-objective hyperparameter optimization.
result MO ASHA enables scalable multi-objective hyperparameter optimization.
A new algorithm reduces medoid computation time by 4-5 orders of magnitude.
problem Computing the medoid of a set of points efficiently.
method Adapting multi-armed bandit sampling strategy to reduce distance computations.
result Significant gains in both number of distance computations and wall clock time.
Derives a size premium from automated market makers in decentralized AI subnets.
problem Determining the profitability and risk of decentralized AI subnets.
method Analyzes daily data on 128 subnets, tests the size premium, and calculates transaction costs.
result The size premium is reduced by a halving of token emissions but remains profitable only below a certain asset threshold.
Meta-learning improves hyperparameter tuning for XGBoost.
problem Improving hyperparameter tuning for XGBoost models.
method Proposed MeSH algorithm using meta-regressors to guide hyperparameter selection.
result MeSH often finds superior hyperparameter configurations compared to SH and random search.
RAMPART ranks top-k features more accurately than existing methods.
problem Accurate ranking of important features in machine learning.
method Adaptive sequential halving strategy combined with ensembling techniques.
result RAMPART achieves the correct top-k ranking with high probability.
In this article, we show the existence of conjugations on many simply-connected spin 6-manifolds with free integral cohomology. In a certain class the only condition on X^6 to admit a conjugation with fixed point set M^3 is the obvious one: the existence of a degree-halving ring isomorphism between the Z_2-cohomologies…
Algorithm clusters items by sequentially selecting features, minimizing observations.
problem Clustering items based on bandit feedback with many features.
method Sequential Halving algorithm for feature selection.
result Accurate recovery of item partition with minimal observations.
Bayesian optimization outperforms other methods in hyperparameter tuning for reinforcement learning.
problem Finding optimal hyperparameters that generalize across random seeds in reinforcement learning.
method Benchmarked Successive Halving, Random Search, and Bayesian Optimization with and without repetitions on PPO2 algorithms for Cartpole and Inverted Pendulum tasks.
result Bayesian optimization with noise robust acquisition function is the best choice.
This paper proposes a new AED framework for multi-metric experiments with fixed budget.
problem Statistical power challenges in testing multiple metrics simultaneously.
method Two-phase structure: adaptive exploration followed by validation. SHRVar algorithm with relative-variance-based sampling.
result Achieves provable error probability that decreases exponentially.
It is shown that the determinant line bundle associated to a family of Dirac operators over a closed partitioned manifold has a canonical Hermitian metric with compatible connection whose curvature satisfies an additivity formula with contributions from the families of Dirac operators over the two halves. This curvatur…
BOSH improves decision-based attacks by optimizing solution paths.
problem Generating optimal adversarial examples for decision-based attacks.
method BOSH-attack uses Bayesian Optimization and Successive Halving to explore solution paths.
result BOSH converges to better solutions with fewer queries.
In their article "The shape of hyperbolic Dehn surgery space," Hodgson and Kerckhoff proved a powerful theorem, half of which they used to make Thurston's Dehn surgery theorem effective. The calculations derived here use both halves of Hodgson and Kerckhoff's theorem to give bounds leading towards a practical algorithm…
New bounds for average graph distance using curvature and centrality.
problem Finding bounds for average graph distance.
method Using weighted average Ollivier curvature with edge betweenness centrality.
result Equality in bounds achieved for specific reflective graphs.
Seesaw optimizes training by balancing learning rate and batch size, accelerating model pretraining.
problem Optimizing training efficiency for large language models with adaptive optimizers.
method Develops a principled framework for batch-size scheduling, introducing Seesaw which multiplies learning rate by 1/√2 and doubles batch size.
result Empirically, Seesaw reduces wall-clock time by approximately 36% compared to cosine decay, matching theoretical limits.
We construct higher genus Riemann's minimal surfaces properly embedded in the Euclidean space. To do that we glue end by end a Costa-Hoffman-Meeks examples to two halves genus zero Riemann's minimal surfaces. In first we need to perform a deformation of a Costa-Hoffman-Meeks example to prescribe the flux vector along t…
New algorithms improve stopping time for best arm identification.
problem Efficiently identifying the best alternative in experiments.
method Proposed algorithms with exponential-tailed stopping time.
result Proved that some algorithms never stop, leading to new methods.
Khovanov homology extended to 3-manifolds, linking tangles.
problem Extending Khovanov homology to 3-manifolds.
method Cutting 3-manifolds into intervals, constructing type D and A structures, gluing back.
result Recovering Khovanov homology of links in 3-manifolds.
Designs efficient factorial experiments for product design under budget constraints.
problem Designing effective experiments for product design with limited traffic and overlapping experiments.
method Two-stage design: first stage samples and infers performance, second stage selects a final policy.
result The method outperforms one-shot tensor completion and unstructured best-arm benchmarks.
New algorithm reduces hyperparameter search space using group sparsity.
problem Efficient hyperparameter selection in machine learning.
method Modifies Harmonica algorithm with group-sparse recovery and HyperBand.
result Improves over existing methods like Successive Halving and Random Search.
Motivated by recent advance of machine learning using Deep Reinforcement Learning this paper proposes a modified architecture that produces more robust agents and speeds up the training process. Our architecture is based on Asynchronous Advantage Actor-Critic (A3C) algorithm where the total input dimensionality is halv…
There are two halves to RL systems: experience collection time and policy learning time. For a large number of samples in rollouts, experience collection time is the major bottleneck. Thus, it is necessary to speed up the rollout generation time with multi-process architecture support. Our work, dubbed WALL-E, utilizes…
Learning can be seen as approximating an unknown function by interpolating the training data. Kriging offers a solution to this problem based on the prior specification of a kernel. We explore a numerical approximation approach to kernel selection/construction based on the simple premise that a kernel must be good if t…
We derive an optimal policy for adaptively restarting a randomized algorithm, based on observed features of the run-so-far, so as to minimize the expected time required for the algorithm to successfully terminate. Given a suitable Bayesian prior, this result can be used to select the optimal black-box optimization algo…
We introduce a general-purpose conditioning method for neural networks called FiLM: Feature-wise Linear Modulation. FiLM layers influence neural network computation via a simple, feature-wise affine transformation based on conditioning information. We show that FiLM layers are highly effective for visual reasoning - an…
The study bounds the effective diameter of graphs with positive Ollivier curvature.
problem Bounding the effective diameter of graphs with positive Ollivier curvature.
method Introducing reflective graphs and proving discrete Bonnet Myers theorem.
result The effective diameter bound is attained only for specific graphs.
We compute the integer cohomology rings of the ``polygon spaces'' introduced in [Hausmann,Klyachko,Kapovich-Millson]. This is done by embedding them in certain toric varieties; the restriction map on cohomology is surjective and we calculate its kernel using ideas from the theory of Gröbner bases. Since we do not inver…
Arnold introduced invariants , and for generic planar curves. It is known that both and are invariants for generic spherical curves. Applying these invariants to underlying curves of knot diagrams, we can obtain lower bounds for the number of Reidemeister moves for uknotting.…
In earlier studies, the estimation of the volatility of a stock using information on the daily opening, closing, high and low prices has been developed; the additional information in the high and low prices can be incorporated to produce unbiased (or near-unbiased) estimators with substantially lower variance than the …
The two main issues for managing wrong way risk (WWR) for the credit valuation adjustment (CVA, i.e. WW-CVA) are calibration and hedging. Hence we start from a novel model-free worst-case approach based on static hedging of counterparty exposure with liquid options. We say "start from" because we demonstrate that a nai…
Two algorithms improve fitting autoregressive models for big data.
problem Efficiently solving Toeplitz least squares problems for large time series data.
method Applied randomized numerical linear algebra (RandNLA) techniques.
result LSAR algorithm is more robust for real-world time series data.
Pfaffian groupoidsmath.DG
This thesis is about the study of Lie groupoids endowed with a compatible (multiplicative) differential 1-form. The motivation and scope of the present work is to study the geometry of PDEs using the formalism of Lie groupoids and multiplicative forms; as such, ideas from the two theories have to be introduced and expl…
Improved diffusion model generation speed with speculative sampling.
problem Generating samples from computationally expensive diffusion models.
method Extending speculative sampling to diffusion models, using fast draft models for candidate token generation.
result Significant speedup in generation, halving the number of function evaluations.
The interdependent nature of the global economy has become stronger with increases in international trade and investment. We propose a new model to reconstruct the international trade network and associated cost network by maximizing entropy based on local information about inward and outward trade. We show that the tr…
Bob predicts a future observation based on a sample of size one. Alice can draw a sample of any size before issuing her prediction. How much better can she do than Bob? Perhaps surprisingly, under a large class of loss functions, which we refer to as the Cover-Hart family, the best Alice can do is to halve Bob's risk. …
We present a design and implementation of the Thomas algorithm optimized for hardware acceleration on an FPGA, the Thomas Core. The hardware-based algorithm combined with the custom data flow and low level parallelism available in an FPGA reduces the overall complexity from 8N down to 5N serial arithmetic operations, a…
Many iterative procedures in stochastic optimization exhibit a transient phase followed by a stationary phase. During the transient phase the procedure converges towards a region of interest, and during the stationary phase the procedure oscillates in that region, commonly around a single point. In this paper, we devel…
A particular Riemannian metric which originally has been obtained for a well-known coordinate system in the Euclidean 3-space, is shown to specify, in fact, a manifold with boundary. There are two ways to make the manifold complete. One is to identify two halves of the boundary that turns the manifold into Euclidean 3-…
We fully describe the horofunction boundary with the word metric associated with the generating set (i.e the metric arising in the Diestel-Leader graph ). The visual boundary with this metric is a subset of . Although $\partial_\infty L_2…
Low-Rank Thinningstat.ML
New method reduces summary points for datasets while maintaining quality.
problem Thinning datasets to reduce summary points while maintaining quality.
method Low-rank analysis of sub-Gaussian thinning.
result Guarantees high-quality compression for any distribution and kernel.
-vectors of balanced simplicial complexes, flag spheres, and geometric Lefschetz decompositionsmath.CO
New -vectors reveal geometric Lefschetz-like decompositions of flag spheres.
problem Understanding -vectors of balanced simplicial complexes and flag spheres.
method Analyzing -vectors and -vectors of flag spheres and balanced simplicial complexes.
result Found -vectors leading to geometric Lefschetz-like decompositions.
Quantization can improve the execution latency and energy efficiency of neural networks on both commodity GPUs and specialized accelerators. The majority of existing literature focuses on training quantized DNNs, while this work examines the less-studied topic of quantizing a floating-point model without (re)training. …
When applied to training deep neural networks, stochastic gradient descent (SGD) often incurs steady progression phases, interrupted by catastrophic episodes in which loss and gradient norm explode. A possible mitigation of such events is to slow down the learning process. This paper presents a novel approach to contro…
To construct flexible nonlinear predictive distributions, the paper introduces a family of softplus function based regression models that convolve, stack, or combine both operations by convolving countably infinite stacked gamma distributions, whose scales depend on the covariates. Generalizing logistic regression that…
Box Thirding identifies the best arm efficiently under limited samples.
problem Efficiently identifying the best arm with limited sampling.
method Iterative ternary comparison of arms, discarding the weakest and exploring the best.
result Achieves comparable performance to Successive Halving with less predefined parameters.
Constructs minimal capillary cones with specific symmetry and proves their existence and uniqueness.
problem Proves existence and uniqueness of minimal capillary cones with bi-orthogonal symmetry.
method Solves a nonlinear free boundary equation parametrized by the contact angle and uses monotonicity properties.
result Demonstrates that minimizing capillary hypersurfaces can have singularities in codimension 7.
This paper considers a multi-armed bandit game where the number of arms is much larger than the maximum budget and is effectively infinite. We characterize necessary and sufficient conditions on the total budget for an algorithm to return an ε-good arm with probability at least 1 - δ. In such situations, the sample com…