Search

This report details exploratory research undertaken by Newcastle University from November 2022 to May 2023. The aim of this exploratory project was to develop, trial and evaluate approaches to providing indications of how outdoor space is being perceived, used and valued by students in four schools -- to ascertain which approach (or approaches) may be best suited for use as part of Building in Use (BiU) reviews. The report details the three digital methods tested, as well as the enablers and barriers encountered while using them. The digital tools developed can be used individually to obtain limited data, but a combination of several methods is most and can be combined with information from the BiU site visit and staff survey in providing an overall BiU assessment. However, there are barriers to this including financial outlay, support from schools, and the time and effort required to create the data collection tools, even with the templates and prototypes that we detail in the report. We recommend initial discussions between the BiU technical adviser (TA) and school about tools, so that these can be chosen to fit with school procedures, especially relating to online access, and also fulfil particular needs of a specific BiU review. This would form part of a developed relationship between TA and school, beyond the existing one of schools accommodating data collection using pre-specified methods, which would also include feedback from the TA to the school.

This paper examines the application of the Kernel Sum of Squares (KSOS) method for enhancing kernel learning from data, particularly in the context of dynamical systems. Traditional kernel-based methods, despite their theoretical soundness and numerical efficiency, frequently struggle with selecting optimal base kernels and parameter tuning, especially with gradient-based methods prone to local optima. KSOS mitigates these issues by leveraging a global optimization framework with kernel-based surrogate functions, thereby achieving more reliable and precise learning of dynamical systems. Through comprehensive numerical experiments on the Logistic Map, Henon Map, and Lorentz System, KSOS is shown to consistently outperform gradient descent in minimizing the relative-$\rho$ metric and improving kernel accuracy. These results highlight KSOS's effectiveness in predicting the behavior of chaotic dynamical systems, demonstrating its capability to adapt kernels to underlying dynamics and enhance the robustness and predictive power of kernel-based approaches, making it a valuable asset for time series analysis in various scientific fields.

The fourth edition of the "VIPriors: Visual Inductive Priors for Data-Efficient Deep Learning" workshop features two data-impaired challenges. These challenges address the problem of training deep learning models for computer vision tasks with limited data. Participants are limited to training models from scratch using a low number of training samples and are not allowed to use any form of transfer learning. We aim to stimulate the development of novel approaches that incorporate inductive biases to improve the data efficiency of deep learning models. Significant advancements are made compared to the provided baselines, where winning solutions surpass the baselines by a considerable margin in both tasks. As in previous editions, these achievements are primarily attributed to heavy use of data augmentation policies and large model ensembles, though novel prior-based methods seem to contribute more to successful solutions compared to last year. This report highlights the key aspects of the challenges and their outcomes., Comment: arXiv admin note: text overlap with arXiv:2305.19688

We introduce Mixtral 8x7B, a Sparse Mixture of Experts (SMoE) language model. Mixtral has the same architecture as Mistral 7B, with the difference that each layer is composed of 8 feedforward blocks (i.e. experts). For every token, at each layer, a router network selects two experts to process the current state and combine their outputs. Even though each token only sees two experts, the selected experts can be different at each timestep. As a result, each token has access to 47B parameters, but only uses 13B active parameters during inference. Mixtral was trained with a context size of 32k tokens and it outperforms or matches Llama 2 70B and GPT-3.5 across all evaluated benchmarks. In particular, Mixtral vastly outperforms Llama 2 70B on mathematics, code generation, and multilingual benchmarks. We also provide a model fine-tuned to follow instructions, Mixtral 8x7B - Instruct, that surpasses GPT-3.5 Turbo, Claude-2.1, Gemini Pro, and Llama 2 70B - chat model on human benchmarks. Both the base and instruct models are released under the Apache 2.0 license., Comment: See more details at https://mistral.ai/news/mixtral-of-experts/

Barriers in cities, such as administrative boundaries, natural obstacles, railways or major roads are thought to induce segregation. However, the empirical knowledge about this phenomenon is limited. Here, we present a network science framework to assess barriers to urban mobility along their hierarchy, across residential areas and visited amenities. Using GPS mobility data, we construct a network of blocks from the sequence of individual stays in a major European city. A community detection algorithm allows us to partition this network into non-overlapping areas of dense mobility clusters, in which the effect of transportation hubs can be tuned with a parameter. We apply the Symmetric Area Difference index to quantify the overlap between these mobility clusters and the polygons of urban area separated by barriers. Reducing the effect of transportation hubs results in smaller scale mobility clusters that fit better to lower rank administrative or road barriers compared to their higher rank pairs. We find that characteristic urban barriers can replace each other in dividing mobility clusters of different scales. Next, we define the Barrier Crossing Ratio, the fraction of barrier crossings that bridge mobility clusters. The decomposition of this indicator by origins and destinations suggests a significantly higher impact of barriers on those who live closer to the city center and smaller impact on visits to complex amenities. These results contribute to the ongoing discourse on urban segregation, emphasizing the importance of barriers to urban mobility in shaping interactions and mixing.

Social capital may help individuals maintain their mental health. Most empirical work based on small-scale surveys finds that bonding social capital and cohesive social networks are critical for mental well-being, while bridging social capital and diverse networks are considered less important. Here, we link data on antidepressant use of 277,344 small-town residents to a nation-wide online social network. The data enable us to examine how individuals' mental healthcare is related to the spatial characteristics of their social networks including their strong and weak ties. We find that, besides the cohesion of social networks around home, the diversity of connections to distant places is negatively correlated with the probability of antidepressant use. Spatial diversity of social networks is also associated with decreasing dosage in subsequent years. This relationship is independent from the local access to antidepressants and is more prevalent for young individuals. Structural features of spatial social networks are prospectively associated with depression treatment., Comment: 49 pages, 12 figures

Color is a crucial visual cue readily exploited by Convolutional Neural Networks (CNNs) for object recognition. However, CNNs struggle if there is data imbalance between color variations introduced by accidental recording conditions. Color invariance addresses this issue but does so at the cost of removing all color information, which sacrifices discriminative power. In this paper, we propose Color Equivariant Convolutions (CEConvs), a novel deep learning building block that enables shape feature sharing across the color spectrum while retaining important color information. We extend the notion of equivariance from geometric to photometric transformations by incorporating parameter sharing over hue-shifts in a neural network. We demonstrate the benefits of CEConvs in terms of downstream performance to various tasks and improved robustness to color changes, including train-test distribution shifts. Our approach can be seamlessly integrated into existing architectures, such as ResNets, and offers a promising solution for addressing color-based domain shifts in CNNs., Comment: NeurIPS 2023. Code available on https://github.com/Attila94/ceconv

The simplex gradient, a popular numerical differentiation method due to its flexibility, lacks a principled method by which to construct the sample set, specifically the location of function evaluations. Such evaluations, especially from real-world systems, are often noisy and expensive to obtain, making it essential that each evaluation is carefully chosen to reduce cost and increase accuracy. This paper introduces the curvature aligned simplex gradient (CASG), which provably selects the optimal sample set under a mean squared error objective. As CASG requires function-dependent information often not available in practice, we additionally introduce a framework which exploits a history of function evaluations often present in practical applications. Our numerical results, focusing on applications in sensitivity analysis and derivative free optimization, show that our methodology significantly outperforms or matches the performance of the benchmark gradient estimator given by forward differences (FD) which is given exact function-dependent information that is not available in practice. Furthermore, our methodology is comparable to the performance of central differences (CD) that requires twice the number of function evaluations., Comment: 31 Pages, 5 Figures, Submitted to IMA Numerical Analysis

We introduce Mistral 7B v0.1, a 7-billion-parameter language model engineered for superior performance and efficiency. Mistral 7B outperforms Llama 2 13B across all evaluated benchmarks, and Llama 1 34B in reasoning, mathematics, and code generation. Our model leverages grouped-query attention (GQA) for faster inference, coupled with sliding window attention (SWA) to effectively handle sequences of arbitrary length with a reduced inference cost. We also provide a model fine-tuned to follow instructions, Mistral 7B -- Instruct, that surpasses the Llama 2 13B -- Chat model both on human and automated benchmarks. Our models are released under the Apache 2.0 license., Comment: Models and code are available at https://mistral.ai/news/announcing-mistral-7b/

In temporal action localization, given an input video, the goal is to predict which actions it contains, where they begin, and where they end. Training and testing current state-of-the-art deep learning models requires access to large amounts of data and computational power. However, gathering such data is challenging and computational resources might be limited. This work explores and measures how current deep temporal action localization models perform in settings constrained by the amount of data or computational power. We measure data efficiency by training each model on a subset of the training set. We find that TemporalMaxer outperforms other models in data-limited settings. Furthermore, we recommend TriDet when training time is limited. To test the efficiency of the models during inference, we pass videos of different lengths through each model. We find that TemporalMaxer requires the least computational resources, likely due to its simple architecture., Comment: Accepted to the CVEU workshop at ICCV 2023

In this paper we show how Group Equivariant Convolutional Neural Networks use subsampling to learn to break equivariance to their symmetries. We focus on 2D rotations and reflections and investigate the impact of broken equivariance on network performance. We show that a change in the input dimension of a network as small as a single pixel can be enough for commonly used architectures to become approximately equivariant, rather than exactly. We investigate the impact of networks not being exactly equivariant and find that approximately equivariant networks generalise significantly worse to unseen symmetries compared to their exactly equivariant counterparts. However, when the symmetries in the training data are not identical to the symmetries of the network, we find that approximately equivariant networks are able to relax their own equivariant constraints, causing them to match or outperform exactly equivariant networks on common benchmark datasets.

The third edition of the "VIPriors: Visual Inductive Priors for Data-Efficient Deep Learning" workshop featured four data-impaired challenges, focusing on addressing the limitations of data availability in training deep learning models for computer vision tasks. The challenges comprised of four distinct data-impaired tasks, where participants were required to train models from scratch using a reduced number of training samples. The primary objective was to encourage novel approaches that incorporate relevant inductive biases to enhance the data efficiency of deep learning models. To foster creativity and exploration, participants were strictly prohibited from utilizing pre-trained checkpoints and other transfer learning techniques. Significant advancements were made compared to the provided baselines, where winning solutions surpassed the baselines by a considerable margin in all four tasks. These achievements were primarily attributed to the effective utilization of extensive data augmentation policies, model ensembling techniques, and the implementation of data-efficient training methods, including self-supervised representation learning. This report highlights the key aspects of the challenges and their outcomes., Comment: arXiv admin note: text overlap with arXiv:2201.08625

Cities host diverse people and their mixing is the engine of prosperity. In turn, segregation and inequalities are common features of most cities and locations that enable the meeting of people with different socio-economic status are key for urban inclusion. In this study, we adopt the concept of economic complexity to quantify the sophistication of amenity supply at urban locations. We propose that neighborhood complexity and amenity complexity are connected to the ability of locations to attract diverse visitors from various socio-economic backgrounds across the city. We construct the measures of amenity complexity based on the local portfolio of diverse and non-ubiquitous amenities in Budapest, Hungary. Socio-economic mixing at visited third places is investigated by tracing the daily mobility of individuals and by characterizing their status by the real-estate price of their home locations. Results suggest that measures of ubiquity and diversity of amenities do not, but neighborhood complexity and amenity complexity are correlated with the urban centrality of locations. Urban centrality is a strong predictor of socio-economic mixing, but both neighborhood complexity and amenity complexity add further explanatory power to our models. Our work combines urban mobility data with economic complexity thinking to show that the diversity of non-ubiquitous amenities, central locations, and the potentials for socio-economic mixing are interrelated.