There has been growing interest in using photonic processors for performing neural network inference operations; however, these networks are currently trained using standard digital electronics. Here, we propose on-chip training of neural networks enabled by a CMOS-compatible silicon photonic architecture to harness the potential for massively parallel, efficient, and fast data operations. Our scheme employs the direct feedback alignment training algorithm, which trains neural networks using error feedback rather than error backpropagation, and can operate at speeds of trillions of multiply-accumulate (MAC) operations per second while consuming less than one picojoule per MAC operation. The photonic architecture exploits parallelized matrix-vector multiplications using arrays of microring resonators for processing multi-channel analog signals along single waveguide buses to calculate the gradient vector for each neural network layer in situ. We also experimentally demonstrate training deep neural networks with the MNIST dataset using on-chip MAC operation results. Our novel approach for efficient, ultra-fast neural network training showcases photonics as a promising platform for executing AI applications.

One proposed mechanism of language change concerns the role played by second-language (L2) learners in situations of language contact. If sufficiently many L2 speakers are present in a speech community in relation to the number of first-language (L1) speakers, then those features which present a difficulty in L2 acquisition may be prone to disappearing from the language. This paper presents a mathematical account of such contact situations based on a stochastic model of learning and nonlinear population dynamics. The equilibria of a deterministic reduction of the model, describing a mixed population of L1 and L2 speakers, are fully characterized. Whether or not the language changes in response to the introduction of L2 learners turns out to depend on three factors: the overall proportion of L2 learners in the population, the strength of the difficulty speakers face in acquiring the language as an L2, and the language-internal utilities of the competing linguistic variants. These factors are related by a mathematical formula describing a phase transition from retention of the L2-difficult feature to its loss from both speaker populations. This supplies predictions that can be tested against empirical data. Here, the model is evaluated with the help of two case studies, morphological levelling in Afrikaans and the erosion of null subjects in Afro-Peruvian Spanish; the model is found to be broadly in agreement with the historical development in both cases.

Formaldehyde (HCHO) is one one of the most important trace gas in the atmosphere, as it is a pollutant causing respiratory and other diseases. It is also a precursor of tropospheric ozone which damages crops and deteriorates human health. Study of HCHO chemistry and long-term monitoring using satellite data is important from the perspective of human health, food security and air pollution. Dynamic atmospheric chemistry models struggle to simulate atmospheric formaldehyde and often overestimate by up to two times relative to satellite observations and reanalysis. Spatial distribution of modelled HCHO also fail to match satellite observations. Here, we present deep learning approach using a simple super-resolution based convolutional neural network towards simulating fast and reliable atmospheric HCHO. Our approach is an indirect method of HCHO estimation without the need to chemical equations. We find that deep learning outperforms dynamical model simulations which involves complicated atmospheric chemistry representation. Causality establishing the nonlinear relationships of different variables to target formaldehyde is established in our approach by using a variety of precursors from meteorology and chemical reanalysis to target OMI AURA satellite based HCHO predictions. We choose South Asia for testing our implementation as it doesnt have in situ measurements of formaldehyde and there is a need for improved quality data over the region. Moreover, there are spatial and temporal data gaps in the satellite product which can be removed by trustworthy modelling of atmospheric formaldehyde. This study is a novel attempt using computer vision for trustworthy modelling of formaldehyde from remote sensing can lead to cascading societal benefits.

Autonomous Systems (AS) are increasingly proposed, or used, in Safety Critical (SC) applications. Many such systems make use of sophisticated sensor suites and processing to provide scene understanding which informs the AS' decision-making. The sensor processing typically makes use of Machine Learning (ML) and has to work in challenging environments, further the ML-algorithms have known limitations,e.g., the possibility of false-negatives or false-positives in object classification. The well-established safety-analysis methods developed for conventional SC systems are not well-matched to AS, ML, or the sensing systems used by AS. This paper proposes an adaptation of well-established safety-analysis methods to address the specifics of perception-systems for AS, including addressing environmental effects and the potential failure-modes of ML, and provides a rationale for choosing particular sets of guidewords, or prompts, for safety-analysis. It goes on to show how the results of the analysis can be used to inform the design and verification of the AS and illustrates the new method by presenting a partial analysis of a road vehicle. Illustrations in the paper are primarily based on optical sensing, however the paper discusses the applicability of the method to other sensing modalities and its role in a wider safety process addressing the overall capabilities of AS.

An exhaustive study has been conducted to investigate span-based models for the joint entity and relation extraction task. However, these models sample a large number of negative entities and negative relations during the model training, which are essential but result in grossly imbalanced data distributions and in turn cause suboptimal model performance. In order to address the above issues, we propose a two-phase paradigm for the span-based joint entity and relation extraction, which involves classifying the entities and relations in the first phase, and predicting the types of these entities and relations in the second phase. The two-phase paradigm enables our model to significantly reduce the data distribution gap, including the gap between negative entities and other entities, as well as the gap between negative relations and other relations. In addition, we make the first attempt at combining entity type and entity distance as global features, which has proven effective, especially for the relation extraction. Experimental results on several datasets demonstrate that the spanbased joint extraction model augmented with the two-phase paradigm and the global features consistently outperforms previous state-of-the-art span-based models for the joint extraction task, establishing a new standard benchmark. Qualitative and quantitative analyses further validate the effectiveness the proposed paradigm and the global features.

We introduce version 3 of NetKet, the machine learning toolbox for many-body quantum physics. NetKet is built around neural-network quantum states and provides efficient algorithms for their evaluation and optimization. This new version is built on top of JAX, a differentiable programming and accelerated linear algebra framework for the Python programming language. The most significant new feature is the possibility to define arbitrary neural network ans\"atze in pure Python code using the concise notation of machine-learning frameworks, which allows for just-in-time compilation as well as the implicit generation of gradients thanks to automatic differentiation. NetKet 3 also comes with support for GPU and TPU accelerators, advanced support for discrete symmetry groups, chunking to scale up to thousands of degrees of freedom, drivers for quantum dynamics applications, and improved modularity, allowing users to use only parts of the toolbox as a foundation for their own code.

Data owners face increasing liability for how the use of their data could harm under-priviliged communities. Stakeholders would like to identify the characteristics of data that lead to algorithms being biased against any particular demographic groups, for example, defined by their race, gender, age, and/or religion. Specifically, we are interested in identifying subsets of the feature space where the ground truth response function from features to observed outcomes differs across demographic groups. To this end, we propose FORESEE, a FORESt of decision trEEs algorithm, which generates a score that captures how likely an individual's response varies with sensitive attributes. Empirically, we find that our approach allows us to identify the individuals who are most likely to be misclassified by several classifiers, including Random Forest, Logistic Regression, Support Vector Machine, and k-Nearest Neighbors. The advantage of our approach is that it allows stakeholders to characterize risky samples that may contribute to discrimination, as well as, use the FORESEE to estimate the risk of upcoming samples.

In an era of information explosion, recommendation systems play an important role in people's daily life by facilitating content exploration. It is known that user activeness, i.e., number of behaviors, tends to follow a long-tail distribution, where the majority of users are with low activeness. In practice, we observe that tail users suffer from significantly lower-quality recommendation than the head users after joint training. We further identify that a model trained on tail users separately still achieve inferior results due to limited data. Though long-tail distributions are ubiquitous in recommendation systems, improving the recommendation performance on the tail users still remains challenge in both research and industry. Directly applying related methods on long-tail distribution might be at risk of hurting the experience of head users, which is less affordable since a small portion of head users with high activeness contribute a considerate portion of platform revenue. In this paper, we propose a novel approach that significantly improves the recommendation performance of the tail users while achieving at least comparable performance for the head users over the base model. The essence of this approach is a novel Gradient Aggregation technique that learns common knowledge shared by all users into a backbone model, followed by separate plugin prediction networks for the head users and the tail users personalization. As for common knowledge learning, we leverage the backward adjustment from the causality theory for deconfounding the gradient estimation and thus shielding off the backbone training from the confounder, i.e., user activeness. We conduct extensive experiments on two public recommendation benchmark datasets and a large-scale industrial datasets collected from the Alipay platform. Empirical studies validate the rationality and effectiveness of our approach.

This paper deals with the control of laser spot in the context of minimally invasive surgery of the middle ear, e.g., cholesteatoma removal. More precisely, our work is concerned with the exhaustive burring of residual infected cells after primary mechanical resection of the pathological tissues since the latter cannot guarantee the treatment of all the infected tissues, the remaining infected cells cause regeneration of the diseases in 20%-25\-% of cases, which require a second surgery 12-18 months later. To tackle such a complex surgery, we have developed a robotic platform that consists of the combination of a macro-scale system (7 degrees of freedom (DoFs) robotic arm) and a micro-scale flexible system (2 DoFs) which operates inside the middle ear cavity. To be able to treat the residual cholesteatoma regions, we proposed a method to automatically generate optimal laser scanning trajectories inside the regions and between them. The trajectories are tacked using an image-based control scheme. The proposed method and materials were validated experimentally using the lab-made robotic platform. The obtained results in terms of accuracy and behaviour meet perfectly the laser surgery requirements.

Missing data scenarios are very common in ML applications in general and time-series/sequence applications are no exceptions. This paper pertains to a novel Recurrent Neural Network (RNN) based solution for sequence prediction under missing data. Our method is distinct from all existing approaches. It tries to encode the missingness patterns in the data directly without trying to impute data either before or during model building. Our encoding is lossless and achieves compression. It can be employed for both sequence classification and forecasting. We focus on forecasting here in a general context of multi-step prediction in presence of possible exogenous inputs. In particular, we propose novel variants of Encoder-Decoder (Seq2Seq) RNNs for this. The encoder here adopts the above mentioned pattern encoding, while at the decoder which has a different structure, multiple variants are feasible. We demonstrate the utility of our proposed architecture via multiple experiments on both single and multiple sequence (real) data-sets. We consider both scenarios where (i)data is naturally missing and (ii)data is synthetically masked.