Search

We present the history-aware transformer (HAT), a transformer-based model that uses shoppers' purchase history to personalise outfit predictions. The aim of this work is to recommend outfits that are internally coherent while matching an individual shopper's style and taste. To achieve this, we stack two transformer models, one that produces outfit representations and another one that processes the history of purchased outfits for a given shopper. We use these models to score an outfit's compatibility in the context of a shopper's preferences as inferred from their previous purchases. During training, the model learns to discriminate between purchased and random outfits using 3 losses: the focal loss for outfit compatibility typically used in the literature, a contrastive loss to bring closer learned outfit embeddings from a shopper's history, and an adaptive margin loss to facilitate learning from weak negatives. Together, these losses enable the model to make personalised recommendations based on a shopper's purchase history. Our experiments on the IQON3000 and Polyvore datasets show that HAT outperforms strong baselines on the outfit Compatibility Prediction (CP) and the Fill In The Blank (FITB) tasks. The model improves AUC for the CP hard task by 15.7% (IQON3000) and 19.4% (Polyvore) compared to previous SOTA results. It further improves accuracy on the FITB hard task by 6.5% and 9.7%, respectively. We provide ablation studies on the personalisation, constrastive loss, and adaptive margin loss that highlight the importance of these modelling choices.

Classification with rejection emerges as a learning paradigm which allows models to abstain from making predictions. The predominant approach is to alter the supervised learning pipeline by augmenting typical loss functions, letting model rejection incur a lower loss than an incorrect prediction. Instead, we propose a different distributional perspective, where we seek to find an idealized data distribution which maximizes a pretrained model's performance. This can be formalized via the optimization of a loss's risk with a $ \phi$-divergence regularization term. Through this idealized distribution, a rejection decision can be made by utilizing the density ratio between this distribution and the data distribution. We focus on the setting where our $ \phi $-divergences are specified by the family of $ \alpha $-divergence. Our framework is tested empirically over clean and noisy datasets.

Solar cells based on inorganic perovskite CsPbI3 are promising candidates to resolve the challenge of operational stability in the field of perovskite photovoltaics. For stable operation, however, it is crucial to thoroughly understand the extractive and recombinative processes occurring at the interfaces of perovskite and the charge-selective layers. In this study, we focus on the electronic properties of (doped) TiO2 as an electron-selective contact. We show via KPFM that co-doping of TiO2 with Nb(V) and Sn(IV) reduces the materials work function by 270 meV, giving it stronger n-type characteristics compared to Nb(V) mono-doped TiO2. The altered electronic alignment with CsPbI3 translates to enhanced electron extraction, as demonstrated with ssPL, trPL and trSPV in triad. Importantly, we extract crucial parameters, such as the concentration of extracted electrons and the interface hole recombination velocity, from the SPV transients via 2D drift-diffusion simulations. When implementing the co-doped TiO2 into full n-i-p solar cells, the operational stability is enhanced to 32000 h of projected TS80 lifetime. This study provides fundamental understanding of interfacial charge extraction and its correlation with operational stability of perovskite solar cells, which can be transferred to other charge-selective contacts., Comment: 20 pages, 2 figures

Efficient semi-transparent solar cells can trigger the adoption of building integrated photovoltaics. Halide perovskites are particularly suitable in this respect owing to their tunable bandgap. Main drawbacks in the development of transparent perovskite solar cells are the high Voc deficit and the difficulties in depositing thin films over large area substrates, given the low solubility of bromide and chloride precursors. In this work, we develop a 2D and passivation strategies for the high band-gap Br perovskite able to reduce charge recombination and consequently improving the open-circuit voltage. We demonstrate 1cm 2 perovskite solar cells with Voc up to 1.73 V (1.83 eV QFLS) and a PCE of 8.2%. The AVT exceeds 70% by means of a bifacial light management and a record light utilization efficiency of 5.72 is achieved, setting a new standard for transparent photovoltaics. Moreover, we show the high ceiling of our technology towards IoT application due to a bifaciality factor of 87% along with 17% PCE under indoor lighting. Finally, the up-scaling has been demonstrated fabricating 20cm 2 -active area modules with PCE of 7.3% and Voc per cell up to 1.65V.

Providing technologies to communities or domains where training data is scarce or protected e.g., for privacy reasons, is becoming increasingly important. To that end, we generalise methods for unsupervised transfer from multiple input models for structured prediction. We show that the means of aggregating over the input models is critical, and that multiplying marginal probabilities of substructures to obtain high-probability structures for distant supervision is substantially better than taking the union of such structures over the input models, as done in prior work. Testing on 18 languages, we demonstrate that the method works in a cross-lingual setting, considering both dependency parsing and part-of-speech structured prediction problems. Our analyses show that the proposed method produces less noisy labels for the distant supervision.

We extend the definition of the marginal causal effect to the continuous treatment setting and develop a novel characterization of causal bias in the framework of structural causal models. We prove that our derived bias expression is zero if, and only if, the causal effect is identifiable via covariate adjustment. We show that under some restrictions on the structural equations, the causal bias can be estimated efficiently and allows for causal regularization of predictive probabilistic models. We demonstrate the effectiveness of our method for causal bias quantification in various settings where (not) controlling for certain covariates would introduce causal bias.

Cross-lingual transfer is a leading technique for parsing low-resource languages in the absence of explicit supervision. Simple `direct transfer' of a learned model based on a multilingual input encoding has provided a strong benchmark. This paper presents a method for unsupervised cross-lingual transfer that improves over direct transfer systems by using their output as implicit supervision as part of self-training on unlabelled text in the target language. The method assumes minimal resources and provides maximal flexibility by (a) accepting any pre-trained arc-factored dependency parser; (b) assuming no access to source language data; (c) supporting both projective and non-projective parsing; and (d) supporting multi-source transfer. With English as the source language, we show significant improvements over state-of-the-art transfer models on both distant and nearby languages, despite our conceptually simpler approach. We provide analyses of the choice of source languages for multi-source transfer, and the advantage of non-projective parsing. Our code is available online., Comment: Accepted at EACL 2021

Thin (approximately 10 nm) oxide buffer layers grown over lead-halide perovskite device stacks are critical for protecting the perovskite against mechanical and environmental damage. However, the limited perovskite stability restricts the processing methods and temperatures (<=110 C) that can be used to deposit the oxide overlayers, with the latter limiting the electronic properties of the oxides achievable. In this work, we demonstrate an alternative to existing methods that can grow pinhole-free TiOx (x = 2.00+/-0.05) films with the requisite thickness in <1 min without vacuum. This technique is atmospheric pressure chemical vapor deposition (AP-CVD). The rapid but soft deposition enables growth temperatures of >=180 {\deg}C to be used to coat the perovskite. This is >=70 {\deg}C higher than achievable by current methods and results in more conductive TiOx films, boosting solar cell efficiencies by >2%. Likewise, when AP-CVD SnOx (x ~ 2) is grown on perovskites, there is also minimal damage to the perovskite beneath. The SnOx layer is pinhole-free and conformal, which reduces shunting in devices, and increases steady-state efficiencies from 16.5% (no SnOx) to 19.4% (60 nm SnOx), with fill factors reaching 84%. This work shows AP-CVD to be a versatile technique for growing oxides on thermally-sensitive materials., Comment: R.D.R and R.A.J contributed equally. 23 pages. 6 figures

Grounding is crucial for natural language understanding. An important subtask is to understand modified color expressions, such as 'dirty blue'. We present a model of color modifiers that, compared with previous additive models in RGB space, learns more complex transformations. In addition, we present a model that operates in the HSV color space. We show that certain adjectives are better modeled in that space. To account for all modifiers, we train a hard ensemble model that selects a color space depending on the modifier color pair. Experimental results show significant and consistent improvements compared to the state-of-the-art baseline model., Comment: EMNLP 2019 (5 pages + 1 references)

Design and modification of the interfaces, always a critical issue for semiconductor devices, has become the primary tool to harness the full potential of halide perovskite (HaP)-based ones. In particular the outstanding improvements in HaP solar cell performance and stability can be primarily ascribed to a careful choice of the interfacial layout in the layer stack. In this review we describe the unique challenges and opportunities of these approaches (section A). For this purpose, we first elucidate the basic physical and chemical properties of the exposed HaP thin film and crystal surface (section B). We then lay out the energetic alignment processes to adjacent transport and buffer layers (section C) and finally elaborate on the impact of the interface formation on how well/poor a device functions. Based on those sections we then present a road map for the next steps in interfacial design principles for HaP semiconductors (section D).

The process of translation is ambiguous, in that there are typically many valid trans- lations for a given sentence. This gives rise to significant variation in parallel cor- pora, however, most current models of machine translation do not account for this variation, instead treating the prob- lem as a deterministic process. To this end, we present a deep generative model of machine translation which incorporates a chain of latent variables, in order to ac- count for local lexical and syntactic varia- tion in parallel corpora. We provide an in- depth analysis of the pitfalls encountered in variational inference for training deep generative models. Experiments on sev- eral different language pairs demonstrate that the model consistently improves over strong baselines., Comment: Accepted at ACL 2018

Organic-inorganic perovskite solar cells have attracted tremendous attention because of their remarkably high power conversion efficiencies (PCEs). To further improve the device performance, however, it is imperative to obtain fundamental understandings on the photo-response and long-term stability down to the microscopic level. Here, we report the first quantitative nanoscale photoconductivity imaging on two methylammonium lead triiodide (MAPbI3) thin films with different PCEs by light-stimulated microwave impedance microscopy. The intrinsic photo-response is largely uniform across grains and grain boundaries, which is direct evidence on the inherently benign nature of microstructures in the perovskite thin films. In contrast, the carrier mobility and lifetime are strongly affected by bulk properties such as the sample crystallinity. As visualized by the spatial evolution of local photoconductivity, the degradation due to water diffusion through the capping layer begins with the disintegration of large grains rather than the nucleation and propagation from grain boundaries. Our findings provide new insights to improve the electro-optical properties of MAPbI3 thin films towards large-scale commercialization., Comment: 23 pages, 4 figures