Our goal is to improve reliability of Machine Learning (ML) systems deployed in the wild. ML models perform exceedingly well when test examples are similar to train examples. However, real-world applications are required to perform on any distribution of test examples. Current ML systems can fail silently on test examples with distribution shifts. In order to improve reliability of ML models due to covariate or domain shift, we propose algorithms that enable models to: (a) generalize to a larger family of test distributions, (b) evaluate accuracy under distribution shifts, (c) adapt to a target distribution. We study causes of impaired robustness to domain shifts and present algorithms for training domain robust models. A key source of model brittleness is due to domain overfitting, which our new training algorithms suppress and instead encourage domain-general hypotheses. While we improve robustness over standard training methods for certain problem settings, performance of ML systems can still vary drastically with domain shifts. It is crucial for developers and stakeholders to understand model vulnerabilities and operational ranges of input, which could be assessed on the fly during the deployment, albeit at a great cost. Instead, we advocate for proactively estimating accuracy surfaces over any combination of prespecified and interpretable domain shifts for performance forecasting. We present a label-efficient estimation to address estimation over a combinatorial space of domain shifts. Further, when a model's performance on a target domain is found to be poor, traditional approaches adapt the model using the target domain's resources. Standard adaptation methods assume access to sufficient labeled resources, which may be impractical for deployed models. We initiate a study of lightweight adaptation techniques with only unlabeled data resources with a focus on language applications.

Mixed-integer optimisation problems can be computationally challenging. Here, we introduce and analyse two efficient algorithms with a specific sequential design that are aimed at dealing with sampled problems within this class. At each iteration step of both algorithms, we first test the feasibility of a given test solution for each and every constraint associated with the sampled optimisation at hand, while also identifying those constraints that are violated. Subsequently, an optimisation problem is constructed with a constraint set consisting of the current basis -- namely, the smallest set of constraints that fully specifies the current test solution -- as well as constraints related to a limited number of the identified violating samples. We show that both algorithms exhibit finite-time convergence towards the optimal solution. Algorithm 2 features a neural network classifier that notably improves the computational performance compared to Algorithm 1. We quantitatively establish these algorithms' efficacy through three numerical tests: robust optimal power flow, robust unit commitment, and robust random mixed-integer linear program.

Is it possible for machines to think like humans? And if it is, how should we go about teaching them to do so? As early as 1950, Alan Turing stated that we ought to teach machines in the way of teaching a child. Reinforcement learning with human feedback (RLHF) has emerged as a strong candidate toward allowing agents to learn from human feedback in a naturalistic manner. RLHF is distinct from traditional reinforcement learning as it provides feedback from a human teacher in addition to a reward signal. It has been catapulted into public view by multiple high-profile AI applications, including OpenAI's ChatGPT, DeepMind's Sparrow, and Anthropic's Claude. These highly capable chatbots are already overturning our understanding of how AI interacts with humanity. The wide applicability and burgeoning success of RLHF strongly motivate the need to evaluate its social impacts. In light of recent developments, this paper considers an important question: can RLHF be developed and used without negatively affecting human societies? Our objectives are threefold: to provide a systematic study of the social effects of RLHF; to identify key social and ethical issues of RLHF; and to discuss social impacts for stakeholders. Although text-based applications of RLHF have received much attention, it is crucial to consider when evaluating its social implications the diverse range of areas to which it may be deployed. We describe seven primary ways in which RLHF-based technologies will affect society by positively transforming human experiences with AI. This paper ultimately proposes that RLHF has potential to net positively impact areas of misinformation, AI value-alignment, bias, AI access, cross-cultural dialogue, industry, and workforce. As RLHF raises concerns that echo those of existing AI technologies, it will be important for all to be aware and intentional in the adoption of RLHF.

We construct a strong PUF with provable security against ML attacks on both classical and quantum computers. The security is guaranteed by the cryptographic hardness of learning decryption functions of public-key cryptosystems, and the hardness of the learning-with-errors (LWE) problem defined on integer lattices. We call our construction the lattice PUF. We construct lattice PUF with a physically obfuscated key and an LWE decryption function block. To allow deployments in different scenarios, we demonstrate designs with different latency-area trade-offs. A compact design uses a highly serialized LFSR and LWE decryption function, while a latency-optimized design uses an unrolled LFSR and a parallel datapath. In addition to theoretical security guarantee, we evaluate empirical resistance to the various leading ML techniques: the prediction error remains above 49.76% after 1 million training CRPs. The resource-efficient design requires only 45 slices for the PUF logic proper, and 351 slices for a fuzzy extractor. The latency-optimized design achieves a 148X reduction in latency, at a 10X increase in PUF hardware utilization. The mean uniformity of PUF responses is 49.98%, the mean uniqueness is 50.00%, and the mean reliability is 1.26%. ILICON physical unclonable functions (PUFs) are security primitives commonly adopted for device identification, authentication, and cryptographic key generation [38]. A PUF exploits the inherent randomness of CMOS technology to generate an output response for a given input challenge. Weak PUFs, which are also called physically obfuscated keys (POKs) [15], have a limited challenge-response pair (CRP) space. The security of a strong PUF requires the associated CRPs to be unpredictable: given a certain set of known CRPs, it should be hard to predict the unobserved CRPs, even with the most powerful machine learning (ML) based modeling attacks. Engineering an ML resistant strong PUF with low hardware cost has been challenging. Variants of the original arbiter PUF (APUF) with stronger modeling attack resilience, including bistable ring PUF and feedforward APUF, have also been broken via ML attacks [37], [35]. The recent interpose PUF (IPUF) [33] proposal is claimed to have provable ML resistance.

JPEG images can be further compressed to enhance the storage and transmission of large-scale image datasets. Existing learned lossless compressors for RGB images cannot be well transferred to JPEG images due to the distinguishing distribution of DCT coefficients and raw pixels. In this paper, we propose a novel framework for learned lossless compression of JPEG images that achieves end-to-end optimized prediction of the distribution of decoded DCT coefficients. To enable learning in the frequency domain, DCT coefficients are partitioned into groups to utilize implicit local redundancy. An autoencoder-like architecture is designed based on the weight-shared blocks to realize entropy modeling of grouped DCT coefficients and independently compress the priors. We attempt to realize learned lossless compression of JPEG images in the frequency domain. Experimental results demonstrate that the proposed framework achieves superior or comparable performance in comparison to most recent lossless compressors with handcrafted context modeling for JPEG images.

Although the problem of hallucinations in neural machine translation (NMT) has received some attention, research on this highly pathological phenomenon lacks solid ground. Previous work has been limited in several ways: it often resorts to artificial settings where the problem is amplified, it disregards some (common) types of hallucinations, and it does not validate adequacy of detection heuristics. In this paper, we set foundations for the study of NMT hallucinations. First, we work in a natural setting, i.e., in-domain data without artificial noise neither in training nor in inference. Next, we annotate a dataset of over 3.4k sentences indicating different kinds of critical errors and hallucinations. Then, we turn to detection methods and both revisit methods used previously and propose using glass-box uncertainty-based detectors. Overall, we show that for preventive settings, (i) previously used methods are largely inadequate, (ii) sequence log-probability works best and performs on par with reference-based methods. Finally, we propose DeHallucinator, a simple method for alleviating hallucinations at test time that significantly reduces the hallucinatory rate. To ease future research, we release our annotated dataset for WMT18 German-English data, along with the model, training data, and code.

The exponentially large discrete search space in mixed-precision quantization (MPQ) makes it hard to determine the optimal bit-width for each layer. Previous works usually resort to iterative search methods on the training set, which consume hundreds or even thousands of GPU-hours. In this study, we reveal that some unique learnable parameters in quantization, namely the scale factors in the quantizer, can serve as importance indicators of a layer, reflecting the contribution of that layer to the final accuracy at certain bit-widths. These importance indicators naturally perceive the numerical transformation during quantization-aware training, which can precisely provide quantization sensitivity metrics of layers. However, a deep network always contains hundreds of such indicators, and training them one by one would lead to an excessive time cost. To overcome this issue, we propose a joint training scheme that can obtain all indicators at once. It considerably speeds up the indicators training process by parallelizing the original sequential training processes. With these learned importance indicators, we formulate the MPQ search problem as a one-time integer linear programming (ILP) problem. That avoids the iterative search and significantly reduces search time without limiting the bit-width search space. For example, MPQ search on ResNet18 with our indicators takes only 0.06 s, which improves time efficiency exponentially compared to iterative search methods. Also, extensive experiments show our approach can achieve SOTA accuracy on ImageNet for far-ranging models with various constraints (e.g., BitOps, compress rate). Code is available on https://github.com/1hunters/LIMPQ.

Deep neural networks based on layer-stacking architectures have historically suffered from poor inherent interpretability. Meanwhile, symbolic probabilistic models function with clear interpretability, but how to combine them with neural networks to enhance their performance remains to be explored. In this paper, we try to marry these two systems for text classification via a structured language model. We propose a Symbolic-Neural model that can learn to explicitly predict class labels of text spans from a constituency tree without requiring any access to span-level gold labels. As the structured language model learns to predict constituency trees in a self-supervised manner, only raw texts and sentence-level labels are required as training data, which makes it essentially a general constituent-level self-interpretable classification model. Our experiments demonstrate that our approach could achieve good prediction accuracy in downstream tasks. Meanwhile, the predicted span labels are consistent with human rationales to a certain degree.

Nebraska's Hau Chan has been selected to be part of an international research effort focused on the ethical deployment of artificial intelligence. The project is part of a broader collaboration between the National Science Foundation and CSIRO (Australia's national science agency) to fund groundbreaking artificial intelligence research that ultimately solves environmental and societal issues. The University of Nebraska–Lincoln will lead U.S. research efforts in collaboration with the New York-based Rensselaer Polytechnic Institute and the University of New South Wales. The work will concentrate on the development of AI-powered solutions to drought, harmful environmental emissions and infectious diseases. Chan, assistant professor in the School of Computing, will serve as principal investigator on a project that will use AI to determine appropriate allocation of resources such as water, vaccines, medical supplies and non-fossil fuel vehicle stations.

From Cornwall and Wales to Essex, Blackpool and the Norfolk coast, Britain has experienced a flurry of earthquakes in the past month. The biggest – a 3.8 magnitude tremor that struck Wales on February 24 – sparked panic as locals reported their beds started to move and walls shook. One resident in the small Welsh town of Abertillery not far from the epicentre said the quake was so noticeable'it felt like the roof was falling off'. The Welsh quake was preceded by several more including a 1.5 magnitude quake in Cornwall and a 3.8 magnitude event off the coast of Great Yarmouth. Here's all you need to know about the British tremors – including whether recent tectonic activity suggests a'big one' is soon to hit parts of the country.