Learning user representations based on historical behaviors lies at the core of modern recommender systems. Recent advances in sequential recommenders have convincingly demonstrated high capability in extracting effective user representations from the given behavior sequences. Despite significant progress, we argue that solely modeling the observational behaviors sequences may end up with a brittle and unstable system due to the noisy and sparse nature of user interactions logged. In this paper, we propose to learn accurate and robust user representations, which are required to be less sensitive to (attack on) noisy behaviors and trust more on the indispensable ones, by modeling counterfactual data distribution. Specifically, given an observed behavior sequence, the proposed CauseRec framework identifies dispensable and indispensable concepts at both the fine-grained item level and the abstract interest level. CauseRec conditionally samples user concept sequences from the counterfactual data distributions by replacing dispensable and indispensable concepts within the original concept sequence. With user representations obtained from the synthesized user sequences, CauseRec performs contrastive user representation learning by contrasting the counterfactual with the observational. We conduct extensive experiments on real-world public recommendation benchmarks and justify the effectiveness of CauseRec with multiaspects model analysis. The results demonstrate that the proposed CauseRec outperforms state-of-the-art sequential recommenders by learning accurate and robust user representations . CCS CONCEPTS• Information systems → Recommender systems.

Effectively representing users lie at the core of modern recommender systems. Since users' interests naturally exhibit multiple aspects, it is of increasing interest to develop multi-interest frameworks for recommendation, rather than represent each user with an overall embedding. Despite their effectiveness, existing methods solely exploit the encoder (the forward flow) to represent multiple aspects of interests. However, without explicit regularization, the interest embeddings may not be distinct from each other nor semantically reflect representative historical items. Towards this end, we propose the Re4 * framework, which leverages the backward flow to reexamine each interest embedding. Specifically, Re4 encapsulates three backward flows, i.e., 1) Re-contrast, which drives each interest embedding to be distinct from other interests using contrastive learning; 2) Re-attend, which ensures the interest-item correlation estimation in the forward flow to be consistent with the criterion used in final recommendation; and 3) Re-construct, which ensures that each interest embedding can semantically reflect the information of representative items that relate to the corresponding interest. We demonstrate the novel forward-backward multi-interest paradigm on ComiRec, and perform extensive experiments on three real-world datasets. Empirical studies validate that Re4 helps to learn learning distinct and effective multi-interest representations. CCS CONCEPTS• Information systems → Recommender systems.

Self-supervised learning, especially contrastive learning, has made an outstanding contribution to the development of many deep learning research fields. Recently, researchers in the acoustic signal processing field noticed its success and leveraged contrastive learning for better music representation. Typically, existing approaches maximize the similarity between two distorted audio segments sampled from the same music. In other words, they ensure a semantic agreement at the music level. However, those coarse-grained methods neglect some inessential or noisy elements at the frame level, which may be detrimental to the model to learn the effective representation of music. Towards this end, this paper proposes a novel Positive-nEgative frame mask for Music Representation based on the contrastive learning framework, abbreviated as PEMR. Concretely, PEMR incorporates a Positive-Negative Mask Generation module, which leverages transformer blocks to generate frame masks on Log-Mel spectrogram. We can generate self-augmented negative and positive samples by masking important components or inessential components, respectively. We devise a novel contrastive learning objective to accommodate both self-augmented positives/negatives sampled from the same music. We conduct experiments on four public datasets. The experimental results of two music-related downstream tasks, music classification and cover song identification, demonstrate the generalization ability and transferability of music representation learned by PEMR. CCS CONCEPTS• Computing methodologies → Unsupervised learning; Learning latent representations.

For the controlled gasoline engine MR479Q, the crank speed, camshaft position, fuel injection, spark ignition timing signals and their relationships under control mode of group ignition and fuel sequential injection were deeply analyzed, then an electronic control unit (ECU) hardware platform solution based on Freescale 16-bit microcontroller MC9S12XEP100 was given out. Taking advantages of the hardware platform itself, a crank event based fuel injection and spark ignition timing control strategy was proposed to enhance traditional fuel injection and ignition reliability. Fuel pulse width, ignition coil dwell time and spark advance control under different engine operating conditions were then designed in detail respectively. The bench test results show that, the fuel injection and spark ignition timing control signals of ECU are accurate and stable enough under steady operating conditions, even under transient operating conditions when step disturbance exists in throttle opening, the fuel pulse width, dwell time and spark advance are also delivered correctly and reliably. The strategy is feasible enough to accomplish precise control of fuel injection and spark ignition.

Diesel emission fluid (DEF) soaking and urea deposits on selective catalytic reduction (SCR) catalysts are critical issues for real diesel engine NH 3 -SCR systems. To investigate the impact of DEF soaking and urea deposits on SCR catalyst performance, fresh Cu-zeolite catalyst samples were drilled from a full-size SCR catalyst. Those samples were impregnated with DEF solutions and subsequently hydrothermally treated to simulate DEF soaking and urea deposits on real SCR catalysts during diesel engine operations. Their SCR performance was then evaluated in a flow reactor with a four-step test protocol. Test results show that the DEF soaking leached some Cu from the SCR catalysts and slightly reduced their Cu loadings. The loss of Cu and associated metal sites on the catalysts weakened their catalytic oxidation abilities and caused lower NO/NH 3 oxidation and lower high-temperature N 2 O selectivity. Lower Cu loading also made the catalysts less active to the decomposition of surface ammonium nitrates and decreased low-temperature N 2 O selectivity. Cu loss during DEF impregnation released more acid sites on the surface of the catalysts and increased their acidities, and more NH 3 was able to be adsorbed and involved in SCR reactions at medium and high temperatures. Due to lower NH 3 oxidation and higher NH 3 storage, the DEF-impregnated SCR catalyst samples showed higher NO x conversion above 400 °C compared with the non-soaked one. The negative impact of urea deposits during DEF impregnation was not clearly observed, because the high-temperature hydrothermal treatment helped to remove the urea deposits.