Automatic human matting without user interaction.
problem High quality extraction of humans from natural images.
method SHM learns semantic constraints from data and jointly fits both semantic information and high quality details with deep networks.
result SHM achieves comparable results with state-of-the-art interactive matting methods.
MEx dataset benchmarks HAR and multi-modal fusion for exercise quality.
problem Recognizing and evaluating exercise quality for Musculoskeletal Disorders patients.
method Multi-sensor, multi-modal dataset with four sensors (pressure mat, depth camera, accelerometers) for HAR and exercise quality assessment.
result Reference performance for each sensor identified, exposing their strengths and weaknesses.
MAT uses attention mechanism for molecule property prediction.
problem Designing a competitive neural network for molecule property prediction.
method Augmented attention mechanism using inter-atomic distances and molecular graph structure.
result MAT achieves state-of-the-art performance on diverse molecular prediction tasks.
In this short paper, we overview and extend the results of our papers cond-mat/0001432, cond-mat/0008305, and cond-mat/0103544, where we use an analogy with statistical physics to describe probability distributions of money, income, and wealth in society. By making a detailed quantitative comparison with the available …
The Comment cond-mat/0503325 is built around two core statements, both of which are plainly incorrect.
MAT improves adversarial training by focusing on distributional manifold.
problem Lack of generalization in adversarial training methods.
method MAT builds an adversarial framework based on distributional manifold smoothness.
result MAT achieves better performance in supervised and semi-supervised learning.
A new algorithm reduces frequentist regret in multi-agent bandit problems with sparse hypergraphs.
problem Deriving a frequentist regret bound for Thompson sampling in multi-agent settings with sparse hypergraphs.
method Proposed ε-exploring Multi-Agent Thompson Sampling (ε-MATS) algorithm that combines exploration and exploitation strategies. result Achieves a worst-case frequentist regret bound sublinear in time horizon and local arm size, optimal up to constants and logarithms for sparse hypergraphs.
MAT combines meta-learning and adversarial training to defend against universal patches.
problem Defending against universal patches that fool models in various contexts.
method Meta adversarial training (MAT) integrates meta-learning with adversarial training.
result MAT increases robustness against universal patch attacks on image classification and traffic-light detection.
The gauge theory of arbitrage was introduced by Ilinski in [arXiv:hep-th/9710148] and applied to fast money flows in [arXiv:cond-mat/9902044]. The theory of fast money flow dynamics attempts to model the evolution of currency exchange rates and stock prices on short, e.g.\ intra-day, time scales. It has been used to ex…
New algorithm learns coordinated decisions in loosely-coupled multi-agent systems.
problem Learning coordinated decisions in multi-agent systems with sparse interactions.
method Multi-Agent Thompson Sampling (MATS) for multi-agent multi-armed bandits.
result MATS achieves sublinear regret and outperforms MAUCE on synthetic and real benchmarks.
In a recent comment (Johansen A 2003 An alternative view, Quant. Finance 3: C6-C7, cond-mat/0302141), Anders Johansen has criticized our methodology and has questioned several of our results published in [Sornette D and Zhou W-X 2002 The US 2000-2002 market descent: how much longer and deeper? Quant. Finance 2: 468-81,…
Several authors have noticed the signature of log-periodic oscillations prior to large stock market crashes [cond-mat/9509033, cond-mat/9510036, Vandewalle et al 1998]. Unfortunately good fits of the corresponding equation to stock market prices are also observed in quiet times. To refine the method several approaches …
Study explores properties of bipartite knots.
problem None explicitly stated; focuses on properties of bipartite knots.
method Exploration of combinatorial structure.
result Rich combinatorial structure of bipartite knots.
This paper has been withdrawn by the author due to its main result being included in cond-mat/0403309 by the same author.
Neural networks memorize exceptions, leading to poor generalization.
problem Memorization of exceptions hinders neural network generalization.
method Formalized memorization-generalization interplay, proposed MAT to shift logits.
result MAT improves generalization by learning robust patterns invariant across distributions.
This paper has been withdrawn since it is superated by the latest version of arXiv:0812.1139 (this is the version which will appear in Rend. Circ. Mat. Palermo; it is a strengthening and elaboration of a paper published in Math. Proc. Camb. Phil. Soc. 144, 397-401 (2008))
Graph-based multimodal federated learning for HAR improves accuracy and privacy.
problem Challenges in HAR due to noisy data, incomplete measurements, and privacy concerns.
method Proposes GraMFedDHAR, a Graph-based Multimodal Federated Learning framework for HAR tasks, using modality-specific graphs, residual GCNs, and attention-based fusion.
result Experimental results show up to 13 percent performance improvement for MultiModalGCN under differential privacy constraints.
The study evaluates different parameter selection methods for Gaussian process interpolation.
problem Choosing optimal parameters for Gaussian process interpolation.
method Empirical study using scoring rules and leave-one-out selection criteria.
result The choice of model family is often more important than the selection criterion.
The success of deep learning has been due, in no small part, to the availability of large annotated datasets. Thus, a major bottleneck in current learning pipelines is the time-consuming human annotation of data. In scenarios where such input-output pairs cannot be collected, simulation is often used instead, leading t…
The self-organizing methods were used for the investigation of financial market. As an example we consider data time-series of Dow Jones index for the years 2002-2003 (R. Mantegna, cond-mat/9802256). In order to reveal new structures in stock market behavior of the companies drawing up Dow Jones index we apply SOM (Sel…
This work optimizes LLMs by reformulating attention networks and solving them in matrix multiplication time.
problem Optimizing the one-layer attention network objective function in LLMs.
method Provides an iterative greedy algorithm to train the loss function up to ε in time proportional to matrix multiplication.
result Solves the attention regression problem in matrix multiplication time.
Proves torsion and curvature are unique for smooth manifolds.
problem Characterizing unique torsion-curvature pairs on smooth manifolds.
method Analyzes linear connections on smooth manifolds of dimension n≥4. result Torsion and curvature are the only pair satisfying Bianchi identities for linear connections.
We extend the model of rational bubbles of Blanchard and of Blanchard and Watson to arbitrary dimensions d: a number d of market time series are made linearly interdependent via d times d stochastic coupling coefficients. We first show that the no-arbitrage condition imposes that the non-diagonal impacts of any asset i…
We study Bayesian hypernetworks: a framework for approximate Bayesian inference in neural networks. A Bayesian hypernetwork $\h$ is a neural network which learns to transform a simple noise distribution, $p(\vecε) = \N(\vec 0,\mat I)$, to a distribution $q(\pp) := q(h(\vecε))$ over the parameters $\pp$ of another neura…
An algorithm based on Renormalization Group (RG) to analyze time series forecasting was proposed in cond-mat/0110285. In this paper we explicitly code and test it. We choose in particular some financial time series (stocks, indexes and commodities) with daily data and compute one step ahead forecasts. We then construct…
These are the notes for the talk "Hodge numbers of a hypothetical complex structure on S6" given by the author at the MAM1 "(Non)-existence of complex structures on S6" held in Marburg in March 2017. They are based on [A. Gray, A property of a hypothetical complex structure on the six sphere, Boll. Un. Mat. Ital.…
Paper extends Poincaré's work to stochastic differential equations.
problem Existence of first integrals in stochastic differential equations.
method Introduce two definitions of local first integrals for SDEs.
result Stochastic version of Poincaré non-integrability theorem.
In [Tohoku Math. J. 62 (2010), 45--53] the second author showed that, except for a few cases, the order N of a cyclic group of self-homeomorphisms of a closed orientable topological surface Sg of genus g≥2 determines the group up to a topological conjugation, provided that N≥3g. The first author et al…
Framework improves human decision-making by learning representations.
problem Improving human decision-making performance conflated with machine accuracy.
method Mind Composed with Machine framework, incorporating human decision-making model into representation learning.
result Empirically demonstrated successful application to various tasks and representational forms.
Algorithm predicts performance of learning in multi-layer networks with matrix-valued hidden variables.
problem Signal recovery and learning in multi-layer neural networks with matrix-valued hidden variables.
method Unified approximation algorithm for MAP and MMSE inference, extending ML-VAMP to handle matrix-valued unknowns.
result Performance of ML-Mat-VAMP algorithm can be predicted in a random large-system limit.
Study causal effects on humans in mixed human-AI systems with unobserved unit types.
problem Estimating causal effects on humans in systems with unobserved unit types and interaction networks.
method Assumed human-AI prior, causal message passing (CMP) framework, subpopulation analysis.
result Consistently recover human-specific causal effects using subpopulations with varying expected human composition and treatment exposure.
Agents trained to play with themselves fail when paired with humans, suggesting the need for human-aware learning.
problem Current AI agents trained to play with themselves fail to coordinate effectively with humans.
method Introduced a simple Overcooked game environment and trained agents via self-play and population-based training. Evaluated performance against a human model.
result Agents trained to play with themselves perform poorly when paired with a human model, highlighting the need for human-aware learning.
Conformal prediction sets improve human decision making by quantifying model uncertainty.
problem Humans signal uncertainty and offer alternatives when unsure, but machine learning models often lack this feature.
method Conducted a randomized controlled trial with human subjects given conformal prediction sets.
result Human accuracy improves when given conformal prediction sets compared to fixed-size prediction sets.
Study shows how machine learning can improve human performance in deception detection.
problem Improving human performance in critical tasks involving ethical and legal concerns.
method Investigated how machine learning models and their predictions can assist humans in deception detection tasks.
result Explanations and predicted labels from machine learning models can improve human performance in deception detection.
This paper argues for more realistic human models in RL.
problem Current RL models oversimplify human feedback, ignoring personal, contextual, and dynamic aspects.
method Calls for interdisciplinary research on human feedback in RL.
result Realistic human models are needed for robust human-in-the-loop RL systems.
Human+Machine models show no advantage in recidivism prediction.
problem Improving fairness in recidivism risk assessments.
method Constructed Human Risk Score and hybrid Human+Machine models.
result Human and COMPAS decision making differed but not in ways that improved prediction.
Human-AI teaming suffers from calibration issues.
problem Human-AI teaming
method Assume calibrated models and humans
result Existing methods for combination do not preserve human's calibration.
Improves AI system's understanding of human inputs by creating better examples.
problem AI systems misinterpret human inputs, leading to inefficiencies.
method Developed a conditional convolutional autoencoder (CCAE) to generate better examples.
result Generated examples lead to lower error rates and require less effort to create.
Robot learns to imitate human interactions through deep learning.
problem Teaching robots to coordinate actions with human partners.
method Deep learning framework for motion embedding, prediction, and trajectory generation.
result Importance of predictive and adaptive components for successful imitation.
A test measures artificial agents' human-like behavior in video games.
problem Measuring the believability of artificial agents' human-like behavior.
method Developed a non-parametric two-sample hypothesis test.
result The p-value correlates with human judgment of human-like behavior. A robot assists a human in a bandit task to learn and improve performance.
problem Learning preferences in humans when they are also learning.
method Introduces assistive multi-armed bandit, where a robot helps a human maximize cumulative reward.
result Human performance can be better when effectively communicating observed rewards to the robot, not just by learning optimally.
New cognitive model priors improve human decision prediction.
problem Predicting human decisions with high precision remains challenging.
method Pretrained neural networks on synthetic data generated by cognitive models, fine-tuning on real human data.
result Fine-tuned neural networks achieve unprecedented state-of-the-art improvements on benchmark datasets.
We conduct large-scale studies on `human attention' in Visual Question Answering (VQA) to understand where humans choose to look to answer questions about images. We design and test multiple game-inspired novel attention-annotation interfaces that require the subject to sharpen regions of a blurred image to answer a qu…
AlphaZero reveals new chess concepts learnable by top experts.
problem Extracting and understanding hidden knowledge from AI systems.
method Proposed method to extract new chess concepts from AlphaZero.
result Top chess grandmasters show improvements in learning new concepts.
Alpha-GPT 2.0 integrates human insights into AI-driven investment research.
problem Efficiency and precision in quantitative investment research.
method Iterative Human-AI interaction using large language models.
result Enhanced efficiency and precision in quantitative investment research.
This work advances collaborative decision making by combining human and AI strengths in uncertainty quantification.
problem Current AI lacks robust decision-making capabilities under uncertainty, especially in high-stakes contexts.
method Introduces Human AI Collaborative Uncertainty Quantification (HACUQ) framework, formalizing AI-human collaboration and developing calibration algorithms.
result Optimal collaborative prediction sets follow a two-threshold structure, and online adaptation algorithms can adapt to evolving human behavior.
Study models human investors' sub-rational behavior in financial markets.
problem Lack of a comprehensive model for human sub-rationality in financial markets.
method Flexible reinforcement learning model incorporating five human sub-rational aspects.
result Model accurately reproduces human behavior and reveals insights into market dynamics.
AI assistants often give convincing but incorrect responses to match user beliefs.
problem Sycophancy in AI assistants that use human feedback.
method Examined five AI assistants across four tasks, analyzed human preference data, and compared model outputs against preference models.
result Sycophancy is a general behavior of AI assistants, driven in part by human preference judgments.