Low-Rank Adaptation (LoRA) has emerged as a highly efficient framework for finetuning the weights of large foundation models, and has become the go-to method for data-driven customization of LLMs. Despite the promise of highly customized behaviors and capabilities, switching between relevant LoRAs in a multiturn setting is inefficient, as the key-value (KV) cache of the entire turn history must be recomputed with the LoRA weights before generation can begin. To address this problem, we propose Activated LoRA (aLoRA), an adapter architecture which modifies the LoRA framework to only adapt weights for the tokens in the sequence after the aLoRA is invoked. This change crucially allows aLoRA to accept the base model's KV cache of the input string, meaning that aLoRA can be instantly activated whenever needed in a chain without recomputing the prior keys and values. This enables building what we call intrinsics, i.e. specialized models invoked to perform well-defined operations on portions of an input chain or conversation that otherwise uses the base model by default. We train a set of aLoRA-based intrinsics models, demonstrating competitive accuracy with standard LoRA while significantly improving inference efficiency. We contributed our Activated LoRA implementation to the Huggingface PEFT library.

The proliferation of Large Language Models (LLMs) in diverse applications underscores the pressing need for robust security measures to thwart potential jailbreak attacks. These attacks exploit vulnerabilities within LLMs, endanger data integrity and user privacy. Guardrails serve as crucial protective mechanisms against such threats, but existing models often fall short in terms of both detection accuracy, and computational efficiency. This paper advocates for the significance of jailbreak attack prevention on LLMs, and emphasises the role of input guardrails in safeguarding these models. We introduce MoJE (Mixture of Jailbreak Expert), a novel guardrail architecture designed to surpass current limitations in existing state-of-the-art guardrails. By employing simple linguistic statistical techniques, MoJE excels in detecting jailbreak attacks while maintaining minimal computational overhead during model inference. Through rigorous experimentation, MoJE demonstrates superior performance capable of detecting 90% of the attacks without compromising benign prompts, enhancing LLMs security against jailbreak attacks.

Myles Foley, Ambrish Rawat, Taesung Lee, Yufang Hou, Gabriele Picco, Giulio Zizzo. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

The evolving complexity of modern IT infrastructures has paved the way for malicious actors to exploit a wide array of vulnerabilities that can compromise the integrity of these systems. Monitoring complex IT systems is expensive and often requires dedicated infrastructure for deploying Intrusion and/or Anomaly Detection Systems. Moreover, ML-based solutions need large training sets, which add to the overall cost. To tackle these challenges we present INTELLECT, a novel approach to Intrusion and/or Anomaly Detection System, which leverages Federated Learning and model pruning techniques to cooperatively train high-accuracy models using distributed datasets and derive a fleet of lightweight models, which can be deployed without incurring additional costs for dedicated infrastructure. INTELLECT expands on the state-of-the-art techniques for feature selection, model pruning, and model distillation to create an interconnected pipeline. We empirically demonstrate the effectiveness of the methodology on benchmark datasets, and we present guidelines for the deployment in production systems.

Optimizing a machine learning (ML) pipeline has been an important topic of AI and ML. Despite recent progress, pipeline optimization remains a challenging problem, due to potentially many combinations to consider as well as slow training and validation. We present the BLDS algorithm for optimized algorithm selection (ML operations) in a fixed ML pipeline structure. BLDS performs multi-fidelity optimization for selecting ML algorithms trained with smaller computational overhead, while controlling its pipeline search based on multi-armed bandit and limited discrepancy search. Our experiments on well-known classification benchmarks show that BLDS is superior to competing algorithms. We also combine BLDS with hyperparameter optimization, empirically showing the advantage of BLDS.

With the lowering costs of sensors, high-volume and high-velocity data are increasingly being generated and analyzed, especially in IoT domains like energy and smart homes. Consequently, applications that require accurate short-term forecasts and predictions are also steadily increasing. In this paper, we provide an overview of a novel end-to-end platform that provides efficient ingestion, compression, transfer, query processing, and machine learning-based analytics for high-frequency and high-volume time series from IoT. The performance of the platform is evaluated using real-world dataset from RES installations. The results show the importance of high-frequency analytics and the surprisingly positive impact of error bounded lossy compression on machine learning in the form of AutoML. For example, when detecting yaw misalignments in wind turbines, an improvement of 9% in accuracy was observed for AutoML models on lossy compressed data compared to the current industry standard of 10-minute aggregated data. Thus, these small-scale experiments show the potential of the platform, and larger pilots are planned.

AutoML automatically selects, composes and parameterizes machine learning algorithms into a workflow or pipeline of operations that aims at maximizing performance on a given dataset. Although current methods for AutoML achieved impressive results they mostly concentrate on optimizing fixed linear workflows. In this paper, we take a different approach and focus on generating and optimizing pipelines of complex directed acyclic graph shapes. These complex pipeline structure may lead to discovering hidden features and thus boost performance considerably. We explore the power of heuristic search and context-free grammars to search and optimize these kinds of pipelines. Experiments on various benchmark datasets show that our approach is highly competitive and often outperforms existing AutoML systems.

Data science is labor-intensive and human experts are scarce but heavily involved in every aspect of it. This makes data science time consuming and restricted to experts with the resulting quality heavily dependent on their experience and skills. To make data science more accessible and scalable, we need its democratization. Automated Data Science (AutoDS) is aimed towards that goal and is emerging as an important research and business topic. We introduce and define the AutoDS challenge, followed by a proposal of a general AutoDS framework that covers existing approaches but also provides guidance for the development of new methods. We categorize and review the existing literature from multiple aspects of the problem setup and employed techniques. Then we provide several views on how AI could succeed in automating end-to-end AutoDS. We hope this survey can serve as insightful guideline for the AutoDS field and provide inspiration for future research.

The growing interest in both the automation of machine learning and deep learning has inevitably led to the development of a wide variety of methods to automate deep learning. The choice of network architecture has proven critical, and many improvements in deep learning are due to new structuring of it. However, deep learning techniques are computationally intensive and their use requires a high level of domain knowledge. Even a partial automation of this process therefore helps to make deep learning more accessible for everyone. In this tutorial we present a uniform formalism that enables different methods to be categorized and compare the different approaches in terms of their performance. We achieve this through a comprehensive discussion of the commonly used architecture search spaces and architecture optimization algorithms based on reinforcement learning and evolutionary algorithms as well as approaches that include surrogate and one-shot models. In addition, we discuss approaches to accelerate the search for neural architectures based on early termination and transfer learning and address the new research directions, which include constrained and multi-objective architecture search as well as the automated search for data augmentation, optimizers, and activation functions.

Generative Adversarial Networks (GANs) have become a widely popular framework for generative modelling of high-dimensional datasets. However their training is well-known to be difficult. This work presents a rigorous statistical analysis of GANs providing straight-forward explanations for common training pathologies such as vanishing gradients. Furthermore, it proposes a new training objective, Kernel GANs and demonstrates its practical effectiveness on large-scale real-world data sets. A key element in the analysis is the distinction between training with respect to the (unknown) data distribution, and its empirical counterpart. To overcome issues in GAN training, we pursue the idea of smoothing the Jensen-Shannon Divergence (JSD) by incorporating noise in the input distributions of the discriminator. As we show, this effectively leads to an empirical version of the JSD in which the true and the generator densities are replaced by kernel density estimates, which leads to Kernel GANs.