Search

Generative models have enabled easy creation and generation of images of all kinds given a single prompt. However, this has also raised ethical concerns about what is an actual piece of content created by humans or cameras compared to model-generated content like images or videos. Watermarking data generated by modern generative models is a popular method to provide information on the source of the content. The goal is for all generated images to conceal an invisible watermark, allowing for future detection or identification. The Stable Signature finetunes the decoder of Latent Diffusion Models such that a unique watermark is rooted in any image produced by the decoder. In this paper, we present a novel adversarial fine-tuning attack that disrupts the model's ability to embed the intended watermark, exposing a significant vulnerability in existing watermarking methods. To address this, we further propose a tamper-resistant fine-tuning algorithm inspired by methods developed for large language models, tailored to the specific requirements of watermarking in LDMs. Our findings emphasize the importance of anticipating and defending against potential vulnerabilities in generative systems.

The Needle-in-a-haystack (NIAH) test is a general task used to assess language models' (LMs') abilities to recall particular information from long input context. This framework however does not provide a means of analyzing what factors, beyond context length, contribute to LMs' abilities or inabilities to separate and recall needles from their haystacks. To provide a systematic means of assessing what features contribute to LMs' NIAH capabilities, we developed a synthetic benchmark called DENIAHL (Data-oriented Evaluation of NIAH for LLM's). Our work expands on previous NIAH studies by ablating NIAH features beyond typical context length including data type, size, and patterns. We find stark differences between GPT-3.5 and LLaMA 2-7B's performance on DENIAHL, and drops in recall performance when features like item size are increased, and to some degree when data type is changed from numbers to letters. This has implications for increasingly large context models, demonstrating factors beyond item-number impact NIAH capabilities.

Uncertainty quantification (UQ) is the process of systematically determining and characterizing the degree of confidence in computational model predictions. In the context of systems biology, especially with dynamic models, UQ is crucial because it addresses the challenges posed by nonlinearity and parameter sensitivity, allowing us to properly understand and extrapolate the behavior of complex biological systems. Here, we focus on dynamic models represented by deterministic nonlinear ordinary differential equations. Many current UQ approaches in this field rely on Bayesian statistical methods. While powerful, these methods often require strong prior specifications and make parametric assumptions that may not always hold in biological systems. Additionally, these methods face challenges in domains where sample sizes are limited, and statistical inference becomes constrained, with computational speed being a bottleneck in large models of biological systems. As an alternative, we propose the use of conformal inference methods, introducing two novel algorithms that, in some instances, offer non-asymptotic guarantees, enhancing robustness and scalability across various applications. We demonstrate the efficacy of our proposed algorithms through several scenarios, highlighting their advantages over traditional Bayesian approaches. The proposed methods show promising results for diverse biological data structures and scenarios, offering a general framework to quantify uncertainty for dynamic models of biological systems.The software for the methodology and the reproduction of the results is available at https://zenodo.org/doi/10.5281/zenodo.13644870.

PhisNet is a cutting-edge web application designed to detect phishing websites using advanced machine learning. It aims to help individuals and organizations identify and prevent phishing attacks through a robust AI framework. PhisNet utilizes Python to apply various machine learning algorithms and feature extraction techniques for high accuracy and efficiency. The project starts by collecting and preprocessing a comprehensive dataset of URLs, comprising both phishing and legitimate sites. Key features such as URL length, special characters, and domain age are extracted to effectively train the model. Multiple machine learning algorithms, including logistic regression, decision trees, and neural networks, are evaluated to determine the best performance in phishing detection. The model is finely tuned to optimize metrics like accuracy, precision, recall, and the F1 score, ensuring reliable detection of both common and sophisticated phishing tactics. PhisNet's web application is developed using React.js, which allows for client-side rendering and smooth integration with backend services, creating a responsive and user-friendly interface. Users can input URLs and receive immediate predictions with confidence scores, thanks to a robust backend infrastructure that processes data and provides real-time results. The model is deployed using Google Colab and AWS EC2 for their computational power and scalability, ensuring the application remains accessible and functional under varying loads. In summary, PhisNet represents a significant advancement in cybersecurity, showcasing the effective use of machine learning and web development technologies to enhance user security. It empowers users to prevent phishing attacks and highlights AI's potential in transforming cybersecurity.

Living organisms exhibit remarkable adaptations across all scales, from molecules to ecosystems. We believe that many of these adaptations correspond to optimal solutions driven by evolution, training, and underlying physical and chemical laws and constraints. While some argue against such optimality principles due to their potential ambiguity, we propose generalized inverse optimal control to infer them directly from data. This novel approach incorporates multi-criteria optimality, nestedness of objective functions on different scales, the presence of active constraints, the possibility of switches of optimality principles during the observed time horizon, maximization of robustness, and minimization of time as important special cases, as well as uncertainties involved with the mathematical modeling of biological systems. This data-driven approach ensures that optimality principles are not merely theoretical constructs but are firmly rooted in experimental observations. Furthermore, the inferred principles can be used in forward optimal control to predict and manipulate biological systems, with possible applications in bio-medicine, biotechnology, and agriculture. As discussed and illustrated, the well-posed problem formulation and the inference are challenging and require a substantial interdisciplinary effort in the development of theory and robust numerical methods.

We consider the problem of parameter estimation in dynamic systems described by ordinary differential equations. A review of the existing literature emphasizes the need for deterministic global optimization methods due to the nonconvex nature of these problems. Recent works have focused on expanding the capabilities of specialized deterministic global optimization algorithms to handle more complex problems. Despite advancements, current deterministic methods are limited to problems with a maximum of around five state and five decision variables, prompting ongoing efforts to enhance their applicability to practical problems. Our study seeks to assess the effectiveness of state-of-the-art general-purpose global and local solvers in handling realistic-sized problems efficiently, and evaluating their capabilities to cope with the nonconvex nature of the underlying estimation problems.

With the rise of Large Language Models(LLMs), it has become crucial to understand their capabilities and limitations in deciphering and explaining the complex web of causal relationships that language entails. Current methods use either explicit or implicit causal reasoning, yet there is a strong need for a unified approach combining both to tackle a wide array of causal relationships more effectively. This research proposes a novel architecture called Context Aware Reasoning Enhancement with Counterfactual Analysis(CARE CA) framework to enhance causal reasoning and explainability. The proposed framework incorporates an explicit causal detection module with ConceptNet and counterfactual statements, as well as implicit causal detection through LLMs. Our framework goes one step further with a layer of counterfactual explanations to accentuate LLMs understanding of causality. The knowledge from ConceptNet enhances the performance of multiple causal reasoning tasks such as causal discovery, causal identification and counterfactual reasoning. The counterfactual sentences add explicit knowledge of the not caused by scenarios. By combining these powerful modules, our model aims to provide a deeper understanding of causal relationships, enabling enhanced interpretability. Evaluation of benchmark datasets shows improved performance across all metrics, such as accuracy, precision, recall, and F1 scores. We also introduce CausalNet, a new dataset accompanied by our code, to facilitate further research in this domain., Comment: AI Trustworthiness and Risk Assessment for Challenged Contexts (ATRACC) AAAI 2024 Fall Symposium

College students with ADHD respond positively to simple socially assistive robots (SARs) that monitor attention and provide non-verbal feedback, but studies have been done only in brief in-lab sessions. We present an initial design and evaluation of an in-dorm SAR study companion for college students with ADHD. This work represents the introductory stages of an ongoing user-centered, participatory design process. In a three-week within-subjects user study, university students (N=11) with self-reported symptoms of adult ADHD had a SAR study companion in their dorm room for two weeks and a computer-based system for one week. Toward developing SARs for long-term, in-dorm use, we focus on 1) evaluating the usability and desire for SAR study companions by college students with ADHD and 2) collecting participant feedback about the SAR design and functionality. Participants responded positively to the robot; after one week of regular use, 91% (10 of 11) chose to continue using the robot voluntarily in the second week.

This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.

Federated Learning is a training framework that enables multiple participants to collaboratively train a shared model while preserving data privacy and minimizing communication overhead. The heterogeneity of devices and networking resources of the participants delay the training and aggregation in federated learning. This paper proposes a federated learning approach to manoeuvre the heterogeneity among the participants using resource aware clustering. The approach begins with the server gathering information about the devices and networking resources of participants, after which resource aware clustering is performed to determine the optimal number of clusters using Dunn Indices. The mechanism of participant assignment is then introduced, and the expression of communication rounds required for model convergence in each cluster is mathematically derived. Furthermore, a master-slave technique is introduced to improve the performance of the lightweight models in the clusters using knowledge distillation. Finally, experimental evaluations are conducted to verify the feasibility and effectiveness of the approach and to compare it with state-of-the-art techniques., Comment: 13 pages, 4 figures

Following the trend of coefficient bound problems in Geometric Function Theory, in the present paper, we obtain the sharp bound of $|H_3(1)|$ for the class $\mathcal{S}^*$, of starlike functions and $\mathcal{SL}_q^*$, of $q$- starlike functions related with lemniscate of Bernoulli. Bound on the initial class is also an improvement over the existing known bound and the bound on the latter class generalizes the prior known outcome. Further, we determine the extremal functions to prove the sharpness of our results.

Ordinary differential equation models are nowadays widely used for the mechanistic description of biological processes and their temporal evolution. These models typically have many unknown and non-measurable parameters, which have to be determined by fitting the model to experimental data. In order to perform this task, known as parameter estimation or model calibration, the modeller faces challenges such as poor parameter identifiability, lack of sufficiently informative experimental data, and the existence of local minima in the objective function landscape. These issues tend to worsen with larger model sizes, increasing the computational complexity and the number of unknown parameters. An incorrectly calibrated model is problematic because it may result in inaccurate predictions and misleading conclusions. For non-expert users, there are a large number of potential pitfalls. Here, we provide a protocol that guides the user through all the steps involved in the calibration of dynamic models. We illustrate the methodology with two models, and provide all the code required to reproduce the results and perform the same analysis on new models. Our protocol provides practitioners and researchers in biological modelling with a one-stop guide that is at the same time compact and sufficiently comprehensive to cover all aspects of the problem., Comment: 28 pages, 6 figures

In the past several subclasses of starlike functions are defined involving real part and modulus of certain expressions of functions under study, combined by way of an inequality. In the similar fashion, we introduce a new class $\mathcal{S}^*_{q}(\alpha)$, consisting of normalized analytic univalent functions $f$ in the open unit disk $\mathbb{D}$, satisfying $$\RE\left(\dfrac{z f'(z)}{f(z)}\right) \geq \left|1+\dfrac{z f''(z)}{f'(z)} -\dfrac{z f'(z)}{f(z)}-\alpha\right| \quad (0 \leq \alpha <1).$$ Evidently, $\mathcal{S}^*_{q}(\alpha) \subset \mathcal{S}^*$, the class of starlike functions. We first establish $\mathcal{S}^*_{q}(\alpha) \subset \mathcal{S}^*(q_\alpha)$, the class of analytic functions $f$ satisfying $z f'(z)/f(z)\prec q_\alpha(z),$ where $q_\alpha$ is an extremal function. Some necessary and sufficient conditions for functions belonging to these classes are obtained in addition to the inclusion and radius problems. Further, we estimate logarithmic coefficients, inverse coefficients and Fekete-Szeg\"o functional bounds for functions in $ \mathcal{S}^*(q_\alpha)$., Comment: This is the second version(updated) of arXiv:2104.08518, including: added references, q_{\alpha} re-defined for {\alpha}=1/2 (Pg.3,5-7), obtained the extremal function f_{\alpha}(Pg. 5), added radius problems(Sec-5), removed non-sharp bounds of functions results, obtained sharp coefficient bounds

Mechanistic dynamic models allow for a quantitative and systematic interpretation of data and the generation of testable hypotheses. However, these models are often over-parameterized, leading to non-identifiability and non-observability, i.e. the impossibility of inferring their parameters and state variables. The lack of structural identifiability and observability (SIO) compromises a model's ability to make predictions and provide insight. Here we present a methodology, AutoRepar, that corrects SIO deficiencies automatically, yielding reparameterized models that are structurally identifiable and observable. The reparameterization preserves the mechanistic meaning of selected variables, and has the exact same dynamics and input-output mapping as the original model. We implement AutoRepar as an extension of the STRIKE-GOLDD software toolbox for SIO analysis, applying it to several models from the literature to demonstrate its ability to repair their structural deficiencies. AutoRepar increases the applicability of mechanistic models, enabling them to provide reliable information about their parameters and dynamics., Comment: 36 pages, 6 figures

Let $h$ be a non vanishing convex univalent function and $p$ be an analytic function in $\mathbb{D}$. We consider the differential subordination $$\psi_i(p(z), z p'(z)) \prec h(z)$$ with the admissible functions in consideration as $\psi_1:=(\beta p(z)+\gamma)^{-\alpha}\left(\tfrac{(\beta p(z)+\gamma)}{\beta(1-\alpha)}+ z p'(z)\right)$ and $\psi_2:=\tfrac{1}{\sqrt{\gamma \beta}}\arctan\left(\sqrt{\tfrac{\beta}{\gamma}}p^{1-\alpha}(z)\right)+\left(\tfrac{1-\alpha}{\beta p^{2 (1-\alpha)}(z)+\gamma}\right)\tfrac{z p'(z)}{p^{\alpha}(z)}$. The objective of this paper is to find the dominants, preferably the best dominant(say $q$) of the solution of the above differential subordination satisfying $\psi_i(q, n zq'(z))= h(z)$. Further, we show that $\psi_i(q,zq'(z))= h(z)$ is an exact differential equation and $q$ is a convex univalent function in $\mathbb{D}$. In addition, we estimate the sharp lower bound of $\RE p$ for different choices of $h$ and derive a univalence criteria for functions in $\mathcal{H}$(class of analytic normalized functions) as an application to our results.

The recent coronavirus disease (COVID-19) outbreak has dramatically increased the public awareness and appreciation of the utility of dynamic models. At the same time, the dissemination of contradictory model predictions has highlighted their limitations. If some parameters and/or state variables of a model cannot be determined from output measurements, its ability to yield correct insights -- as well as the possibility of controlling the system -- may be compromised. Epidemic dynamics are commonly analysed using compartmental models, and many variations of such models have been used for analysing and predicting the evolution of the COVID-19 pandemic. In this paper we survey the different models proposed in the literature, assembling a list of 36 model structures and assessing their ability to provide reliable information. We address the problem using the control theoretic concepts of structural identifiability and observability. Since some parameters can vary during the course of an epidemic, we consider both the constant and time-varying parameter assumptions. We analyse the structural identifiability and observability of all of the models, considering all plausible choices of outputs and time-varying parameters, which leads us to analyse 255 different model versions. We classify the models according to their structural identifiability and observability under the different assumptions and discuss the implications of the results. We also illustrate with an example several alternative ways of remedying the lack of observability of a model. Our analyses provide guidelines for choosing the most informative model for each purpose, taking into account the available knowledge and measurements., Comment: 25 pages, 7 figures

This paper deals with a special type of Ma-Minda function introduced here with many fascinating facts and interesting applications. It is much akin in all aspects but differs by a condition from its Ma-Minda counterpart. Further, we consider the function:~$1-\log(1+z)$, a special Ma-Minda of the type introduced here, to define a subclass of starlike functions in a similar fashion as we do with Ma-Minda function and is studied for establishing inclusion and radius results. Apart from that, we also deal with the majorization and Bloch function norm problems for the same class. In addition, we obtain the bounds of fourth coefficient:~$a_4$ and second Hankel determinant:~$H_2(2)$ for the functions belonging to a newly defined class using convolution, which generalizes many earlier known results and its association with the special type of Ma-Minda function is also pointed out., Comment: 23 pages, 1 figure

Reproducibility and reusability of the results of data-based modeling studies are essential. Yet, there has been -- so far -- no broadly supported format for the specification of parameter estimation problems in systems biology. Here, we introduce PEtab, a format which facilitates the specification of parameter estimation problems using Systems Biology Markup Language (SBML) models and a set of tab-separated value files describing the observation model and experimental data as well as parameters to be estimated. We already implemented PEtab support into eight well-established model simulation and parameter estimation toolboxes with hundreds of users in total. We provide a Python library for validation and modification of a PEtab problem and currently 20 example parameter estimation problems based on recent studies. Specifications of PEtab, the PEtab Python library, as well as links to examples, and all supporting software tools are available at https://github.com/PEtab-dev/PEtab, a snapshot is available at https://doi.org/10.5281/zenodo.3732958. All original content is available under permissive licenses.

User Interface (UI) prototyping is a necessary step in the early stages of application development. Transforming sketches of a Graphical User Interface (UI) into a coded UI application is an uninspired but time-consuming task performed by a UI designer. An automated system that can replace human efforts for straightforward implementation of UI designs will greatly speed up this procedure. The works that propose such a system primarily focus on using UI wireframes as input rather than hand-drawn sketches. In this paper, we put forward a novel approach wherein we employ a Deep Neural Network that is trained on our custom database of such sketches to detect UI elements in the input sketch. Detection of objects in sketches is a peculiar visual recognition task that requires a specific solution that our deep neural network model attempts to provide. The output from the network is a platform-independent UI representation object. The UI representation object is a dictionary of key-value pairs to represent the UI elements recognized along with their properties. This is further consumed by our UI parser which creates code for different platforms. The intrinsic platform-independence allows the model to create a UI prototype for multiple platforms with single training. This two-step approach without the need for two trained models improves over other methods giving time-efficient results (average time: 129 ms) with good accuracy.

Speech emotion analysis is an important task which further enables several application use cases. The non-verbal sounds within speech utterances also play a pivotal role in emotion analysis in speech. Due to the widespread use of smartphones, it becomes viable to analyze speech commands captured using microphones for emotion understanding by utilizing on-device machine learning models. The non-verbal information includes the environment background sounds describing the type of surroundings, current situation and activities being performed. In this work, we consider both verbal (speech commands) and non-verbal sounds (background noises) within an utterance for emotion analysis in real-life scenarios. We create an indigenous dataset for this task namely "Indian EmoSpeech Command Dataset". It contains keywords with diverse emotions and background sounds, presented to explore new challenges in audio analysis. We exhaustively compare with various baseline models for emotion analysis on speech commands on several performance metrics. We demonstrate that we achieve a significant average gain of 3.3% in top-one score over a subset of speech command dataset for keyword spotting.