Study improves StarCraft bot's strategy selection with partial observations.
problem Selecting effective strategies in real-time strategy games with limited information.
method Utilized full game state information during training to predict opponent strategies.
result Substantial win rate improvements over a fixed-strategy baseline.
AAMDRL uses DRL to manage assets in noisy, changing environments.
problem Learning in noisy, self-adapting environments with sequential data.
method Augmented state information, one-period lag, walk forward analysis.
result AAMDRL outperforms traditional methods in asset management.
New system resists meme coin copy trading bots.
problem Manipulative bots exploit copy trading in illiquid meme coins.
method Multi-agent architecture with LLM and CoT reasoning.
result System outperforms other methods in prediction and economic performance.
The paper analyzes algorithmic trading in cryptocurrency exchanges, finding a profitable strategy involving indirect conversions.
problem Lack of empirical analysis of algorithmic trading in cryptocurrency exchanges.
method Empirical analysis using historical trades data from Binance.
result A triangular arbitrage strategy yields a 14.4 basis point advantage over direct exchanges.
Detects bots in code commits and characterizes their activity.
problem Identifying and separating bot activity from human developers in code commits.
method BIMAN approach using author names, commit messages, files modified, and project associations.
result AUC-ROC value of 0.9 for bot detection.
Modeling DEX liquidity with heterogeneous LPs and MEV bots.
problem Understanding and predicting the dynamics of decentralized cryptocurrency exchanges.
method Mean-field game approach to model liquidity providers' optimal strategies and interactions.
result Calibrated model produces consistent pool exchange rate dynamics and liquidity evolution.
Paper improves recommendation systems by optimizing sequence of items for clicks.
problem Improving recommendation systems robustness against bots and clicks.
method Minimizing pairwise ranking loss over sequences of items, with thresholds to prevent bot influence.
result The proposed algorithms converge and outperform existing methods in various ranking measures.
Recently, Naghi et al. \cite{NAGHI} studied warped product skew CR-submanifold of the form M1×fM⊥ of order 1 of a Kenmotsu manifold Mˉ such that M1=MT×Mθ, where MT, M⊥ and Mθ are invariant, anti-invariant and proper slant submanifolds of Mˉ. The present paper deals wi…
Achilles predicts Gold vs USD with a profitable trading bot.
problem Accurate and consistent prediction of commodity prices.
method LSTM neural network for minute-by-minute predictions.
result $1623.52 profit in 23 days of testing.
Proposes a robust learning strategy for RS over implicit feedback.
problem Robustness in training large-scale Recommender Systems over implicit feedback.
method Minimizes pairwise ranking loss over blocks of non-clicked followed by clicked items, discarding updates for abnormal blocks.
result The approach prevents shifts in item distribution due to bots or few user interactions, improving ranking measures and computation time.
Graph-Hist classifies social media graphs using feature histograms.
problem Classifying large, sparse social media graphs.
method Extracts latent features, bins nodes, and classifies based on multi-channel histograms.
result Improves bot detection in social media graphs.
This paper presents the first two editions of Visual Doom AI Competition, held in 2016 and 2017. The challenge was to create bots that compete in a multi-player deathmatch in a first-person shooter (FPS) game, Doom. The bots had to make their decisions based solely on visual information, i.e., a raw screen buffer. To p…
Rugby-Bot predicts multiple metrics from a single source using fine-grain data.
problem Complexity of sporting events requires multiple metrics for accurate analysis.
method Multi-task learning with fine-grain spatial data and wide-and-deep learning.
result Predictions are consistent and can be in distribution form.
A structure on an almost contact metric manifold is defined as a generalization of well-known cases: Sasakian, quasi-Sasakian, Kenmotsu and cosymplectic. Then we consider a semi-invariant ξ⊥-submanifold of a manifold endowed with such a structure and two topics are studied: the integrability of distributions de…
In this work, we find an equation that relates the Ricci curvature of a riemannian manifold M and the second fundamental forms of two orthogonal foliations of complementary dimensions, F and F⊥, defined on M. Using this equation, we show a sufficient condition for the manifold M to be …
Understanding the influence of hyperparameters on the performance of a machine learning algorithm is an important scientific topic in itself and can help to improve automatic hyperparameter tuning procedures. Unfortunately, experimental meta data for this purpose is still rare. This paper presents a large, free and ope…
An isoparametric hypersurface in unit spheres has two focal submanifolds. Condition A plays a crucial role in the classification theory of isoparametric hypersurfaces in [CCJ07], [Chi16] and [Miy13]. This paper determines CA, the set of points with Condition A in focal submanifolds. It turns out that the points in $…
The study identifies and analyzes cryptocurrency scams on social media.
problem Cryptocurrency fraud, especially 'pump and dump' scams, on social media platforms.
method Combining social media data to identify and predict scams, analyzing bot activity.
result Significant increase in bot activity during pump attempts.
Success with CBs in a customer support bot boosts key metrics.
problem Improving business metrics in customer support bots.
method Used contextual bandits for reinforcement learning.
result Practical solutions to challenges in RL applications.
ScoreGAN uses GANs with IGM to detect bot-generated reviews based on text and scores.
problem Lack of labeled data and bot-generated reviews in fraud review detection.
method ScoreGAN incorporates review text and scores into a GAN framework for data augmentation and detection.
result ScoreGAN outperforms existing methods by 7% and 5% on Yelp and TripAdvisor datasets.
For a Hamiltonian K∈C2(RN×n) and a map u:Ω⊆Rn⟶RN, we consider the supremal functional \[ \label{1} \tag{1} E_\infty (u,Ω) \ :=\ \big\|K(Du)\big\|_{L^\infty(Ω)} . \] The "Euler-Lagrange" PDE associated to \eqref{1} is the quasilinear system \[ \lab…
The paper studies inequalities for warped product submanifolds in nearly Kenmotsu f-manifolds.
problem Investigating inequalities for warped product pseudo slant submanifolds in nearly Kenmotsu f-manifolds.
method Analyzing basic properties and establishing inequalities for the squared norm of the second fundamental form.
result Established general sharp inequalities for the squared norm of the second fundamental form for mixed totally geodesic warped product pseudo slant submanifolds.
The paper studies maps between Riemannian and Kähler manifolds, focusing on Clairaut semi-invariant Riemannian maps.
problem Analyzing maps between Riemannian and Kähler manifolds, particularly Clairaut semi-invariant Riemannian maps.
method Recalled and defined Clairaut semi-invariant Riemannian maps, derived necessary and sufficient conditions for geodesic curves and maps, and explored foliations and product manifolds.
result Necessary and sufficient conditions for various properties of Clairaut semi-invariant Riemannian maps were derived.
Analytic structure found on manifold of idempotent operators.
problem Analyzing structure of idempotent operators in Banach spaces.
method Locally trivial analytic affine-Banach bundle over Grassmann manifold.
result Bi-analytic bijection between tangent bundle and idempotent manifold.
AI agents beat previous best on NetHack, but symbolic bots still outperform.
problem Developing AI agents that can ascend in the NetHack game.
method Used a procedurally generated NetHack Learning Environment for reinforcement learning.
result Symbolic bots outperform neural approaches on NetHack.
Study shows Elo models fail to accurately measure transitive strength in competitive games.
problem Elo models fail to correctly identify the transitive component in real-world competitive games.
method Investigated the challenge of identifying the transitive component in games, proposed an extension of the Elo score.
result Disc ranking system assigns two scores: skill and consistency.
ZAPs rewards users based on activity, making them less susceptible to bots and sybil attacks.
problem Incentive programs in DeFi are vulnerable to bots and sybil operations.
method ZAPs combines percentile normalization, two-layer weighting, and anomaly detection.
result ZAPs reduces adversarial reward capture by 30-90 percent.
This paper sets a lower limit for the size of Weinstein's Lagrangian tubular neighborhoods.
problem Finding the minimum size of Weinstein's Lagrangian tubular neighborhoods.
method Using the curvature tensor and second fundamental form of the submanifold.
result Explicit lower bounds for the radii of tubular neighborhoods are derived.
In this paper, we define a rectifying spacelike curve in the Minkowski space-time E14 as a curve whose position vector always lies in orthogonal complement N⊥ of its principal normal vector field N. In particular, we study the rectifying spacelike curves in E14 and characterize such curves in terms of…
Improved stock price prediction using LSTM with customized loss function and analyst calls.
problem Forecasting stock prices using LSTM neural networks.
method Customized LSTM model with improved loss function, analyst calls integration, and attention units.
result Improved performance of LSTM model over ARIMA for stock price prediction.
The paper studies Riemannian maps with Ricci soliton base manifolds.
problem Analyzing Riemannian maps with specific properties of base manifolds.
method Analyzing Riemannian curvature tensor, Ricci tensor, scalar curvature, and necessary conditions for Ricci soliton leaves.
result Necessary and sufficient conditions for harmonicity and biharmonicity of Riemannian maps.
Framework uses probabilistic programming for physics simulation in games.
problem Efficiently simulate physics in games using probabilistic programming.
method Combines model-free and model-based approaches to improve efficiency.
result Model outperforms model-free or model-based approaches alone.
Improved fairness in pairwise comparisons using MLE with a simple modification.
problem Fairness in pairwise comparisons using MLE is suboptimal.
method Proposed a simple modification to MLE to improve fairness (bias) without sacrificing accuracy.
result Improved rate in bias while maintaining minimax-optimality in mean squared error.
It is proved the non-existence of Hopf hypersurfaces in G2(Cm+2), m≥3, whose normal Jacobi operator is semi-parallel, if the principal curvature of the Reeb vector field is non-vanishing and the component of the Reeb vector field in the maximal quaternionic subbundle D or its orthogonal …
New framework for robustness guarantees in discrete domains.
problem Adversarial robustness in constrained discrete domains.
method Graphical framework for adversarial costs, accommodating complex cost functions.
result Provably minimal adversarial cost and robustness guarantees.
Defines and studies Clairaut Riemannian maps between manifolds and Ricci solitons.
problem Characterizing and analyzing Clairaut Riemannian maps.
method Using geodesic curves, necessary and sufficient conditions for harmonicity and Clairaut maps are derived.
result Necessary conditions for various properties of Clairaut Riemannian maps are established.
The paper studies variations of the Godbillon--Vey invariant for certain foliations.
problem Investigating the Godbillon--Vey invariant for transversely parallelizable foliations.
method Constructing a (2q+1)-form analogous to the Godbillon--Vey class and expressing it in terms of ω and ${f T}$ for a compatible Riemannian metric. result Characterizing critical pairs of $(ω,{f T})$ and finding sufficient conditions for critical foliations.
Data poisoning is an attack on machine learning models wherein the attacker adds examples to the training set to manipulate the behavior of the model at test time. This paper explores poisoning attacks on neural nets. The proposed attacks use "clean-labels"; they don't require the attacker to have any control over the …
New DGA detection models use side info to improve robustness.
problem Adversaries craft domain names to evade DGA detection.
method Train deep learning and RF classifiers using domain name and side info.
result Models using both domain name and side info are more robust.
Let M be a Kähler surface and Σ be a closed symplectic surface which is smoothly immersed in M. Let α be the Kähler angle of Σ in M. We first deduce the Euler-Lagrange equation of the functional L=∫Σcosα1dμ in the class of symplectic surfaces. It is cos3αH=(J(J∇cosα)⊤)⊥, wh…
Proposes a new SGD momentum optimizer using online parabola modeling.
problem Optimizing step size in stochastic gradient descent for neural networks.
method Online parabola modeling in a single direction using normalized momentum vector.
result Improves step size estimation for faster convergence and saddle repulsion.
In this paper we derive the Euler-Lagrange equation of the functional Lβ=∫Σcosβα1dμ, β=−1 in the class of symplectic surfaces. It is cos3αH=β(J(J∇cosα)⊤)⊥, which is an elliptic equation when β≥0. We call such a surface a β-symplectic critical surface. We first st…
We prove, under a certain boundedness condition at infinity on the (Xˉ⊤,Xˉ⊥)-component of the second fundamental form, the vanishing of the essential spectrum of a complete minimal Xˉ-bounded and Xˉ-properly immersed submanifold on a Riemannian manifold endowed with a strongly con…
In this short note, we consider self-similar immersions F:Rn→Rn+k of the Graphic Mean Curvature Flow of higher co-dimension. We show that the following is true: Let F(x)=(x,f(x)),x∈Rn be a graph solution to the soliton equation Hˉ(x)+F⊥(x)=0. Assume…
Improved Zero learning framework wins games like Hex and Havannah.
problem Improving Zero learning for board games with limited resources.
method Fully convolutional structure, global pooling, and checkpointing.
result Won against strong humans in Hex and Havannah.
New method finds better arbitrage opportunities in AMMs.
problem Finding optimal arbitrage trades in multi-token AMMs.
method Closed-form solutions using convex optimisation.
result Better arbitrage opportunities than traditional methods.
In this paper we study the warped product submanifolds of a Lorentzian paracosymplectic manifold and obtain some nonexistence results. We show that a warped product semi-invariant submanifold in the form {M=MT×fM⊥} of Lorentzian paracosymplectic manifold such that the characteristic vector field is n…
Project improves gender-balanced pronoun resolution with BERT.
problem Challenging task in natural language understanding, especially for gendered pronouns.
method BERT-based approach to gender-balanced pronoun resolution.
result Achieved 92% F1 score with lower gender bias.