arXiv research

Efficiently selects top-m designs for various contexts using sequential sampling.

Efficiently allocate budgets for LLM-assisted virtual screening to reduce costs.

The structure set $\ST^{TOP}(M)$ of an $n$-dimensional topological manifold $M$ for $n \geqslant 5$ has a homotopy invariant functorial abelian group structure, by the algebraic version of the Browder-Novikov-Sullivan-Wall surgery theory. An element $(N,f) \in \ST^{TOP}(M)$ is an equivalence class of $n$-dimensional ma…

The top word list, i.e., the top-M words with highest marginal probability in a given topic, is the standard topic representation in topic models. Most of recent automatical topic labeling algorithms and popular topic quality metrics are based on it. However, we find, empirically, words in this type of top word list ar…

We address the M-best-arm identification problem in multi-armed bandits. A player has a limited budget to explore K arms (M<K), and once pulled, each arm yields a reward drawn (independently) from a fixed, unknown distribution. The goal is to find the top M arms in the sense of expected reward. We develop an algorithm …

FOSC-X: An extended framework for extracting multiple optimal flat clusterings from hierarchical cluster trees

We consider combinatorial online learning with subset choices when only relative feedback information from subsets is available, instead of bandit or semi-bandit feedback which is absolute. Specifically, we study two regret minimisation problems over subsets of a finite ground set $[n]$, with subset-wise relative prefe…

We introduce the probably approximately correct (PAC) \emph{Battling-Bandit} problem with the Plackett-Luce (PL) subset choice model--an online learning framework where at each trial the learner chooses a subset of $k$ arms from a fixed set of $n$ arms, and subsequently observes a stochastic feedback indicating prefere…

We study the problem of identifying the top $m$ arms in a multi-armed bandit game. Our proposed solution relies on a new algorithm based on successive rejects of the seemingly bad arms, and successive accepts of the good ones. This algorithmic contribution allows to tackle other multiple identifications settings that w…

The paper develops a theory for speculative decoding acceptance criteria.

Much effort has been directed at algorithms for obtaining the highest probability configuration in a probabilistic random field model known as the maximum a posteriori (MAP) inference problem. In many situations, one could benefit from having not just a single solution, but the top M most probable solutions known as th…

We give an interpretation of Yetter's Invariant of manifolds $M$ in terms of the homotopy type of the function space $TOP(M,B(G))$, where $G$ is a crossed module and $B(G)$ is its classifying space. From this formulation, there follows that Yetter's invariant depends only on the homotopy type of $M$, and the weak homot…

The paper provides robustness guarantees for mode estimation in bandits.

New approach to multi-armed bandit problem aims to maximize highest total reward.

EFS uses LLMs to optimize sparse portfolios by evolving alpha factors.

Efficiently calculates PL model likelihood for partitioned preference data.

We prove that if $M$ is a CW-complex, then the homotopy type of the skeletal filtration of $M$ does not depend on the cell decomposition of $M$ up to wedge products with $n$-disks $D^n$, when the later are given their natural CW-decomposition with unique cells of order 0, $(n-1)$ and $n$; a result resembling J.H.C. Whi…

We prove a \emph{query complexity} lower bound on rank-one principal component analysis (PCA). We consider an oracle model where, given a symmetric matrix $M \in \mathbb{R}^{d \times d}$, an algorithm is allowed to make $T$ \emph{exact} queries of the form $w^{(i)} = Mv^{(i)}$ for $i \in \{1,\dots,T\}$, where $v^{(i)}$…

Bayesian framework identifies dynamical systems from noisy data.

We consider the problem of probably approximately correct (PAC) ranking $n$ items by adaptively eliciting subset-wise preference feedback. At each round, the learner chooses a subset of $k$ items and observes stochastic feedback indicating preference information of the winner (most preferred) item of the chosen subset …

The paper tackles matrix completion in ultra-sparse sampling, improving imputation accuracy.

A framework for faster, better infographic design by non-experts and experts alike.

The paper proposes a method to reliably select design algorithms for machine learning-guided design tasks.

Unified approach to experimental design using interlacing polynomials.

We propose a new low-cost machine-learning-based methodology which assists designers in reducing the gap between the problem and the solution in the design process. Our work applies reinforcement learning (RL) to find the optimal task-oriented design solution through the construction of the design action for each task.…

Design-by-Morphing creates radical airfoil designs without geometric constraints.

MO-PaDGAN generates diverse, high-performance designs with multiple metrics.

MCD automates counterfactual design searches for multi-modal tasks.

PaDGAN generates diverse, high-quality designs with improved performance.

ANN with GA optimizes flexible disc design for lower mass and stress.

New method for mixed-variable GSA improves material design efficiency.

Deep learning models enhance engineering design automation.

Efficient deep learning computing requires algorithm and hardware co-design to enable specialization: we usually need to change the algorithm to reduce memory footprint and improve energy efficiency. However, the extra degree of freedom from the algorithm makes the design space much larger: it's not only about designin…

DAD learns to design experiments quickly, outperforming traditional methods.

Enhances scenario approach for certifying design properties post-design.

MO-PaDGAN improves multi-objective optimization by generating diverse and high-performing designs.

Topology design optimization offers tremendous opportunity in design and manufacturing freedoms by designing and producing a part from the ground-up without a meaningful initial design as required by conventional shape design optimization approaches. Ideally, with adequate problem statements, to formulate and solve the…

Generative thermal design learns optimal shapes using multi-agent reinforcement learning.

Develops a prediction method based on sampling design.

Solving the RNA inverse folding problem is a critical prerequisite to RNA design, an emerging field in bioengineering with a broad range of applications from reaction catalysis to cancer therapy. Although significant progress has been made in developing machine-based inverse RNA folding algorithms, current approaches s…

Paper proposes NASAIC framework for co-designing neural architectures and heterogeneous ASICs.

Adaptive designs achieve strong Neyman regret guarantees for ATE estimation.

iDAD uses neural networks to quickly adapt experiments without likelihoods.

Unified framework for combinatorial and rounding algorithms in experimental design.

New method refines model predictions as design evolves.

A new method achieves optimal uniformity in designs with minimal flexibility.

RID-Noise improves robust design under noisy conditions using neural networks.

The design of multiple experiments is commonly undertaken via suboptimal strategies, such as batch (open-loop) design that omits feedback or greedy (myopic) design that does not account for future effects. This paper introduces new strategies for the optimal design of sequential experiments. First, we rigorously formul…