Explore the vast range of titles available.

Deep neural networks have demonstrated excellent performances in many real-world applications. Unfortunately, they may show Clever Hans-like behaviour (making use of confounding factors within datasets) to achieve high performance. In this work we introduce the novel learning setting of explanatory interactive learning and illustrate its benefits on a plant phenotyping research task. Explanatory interactive learning adds the scientist into the training loop, who interactively revises the original model by providing feedback on its explanations. Our experimental results demonstrate that explanatory interactive learning can help to avoid Clever Hans moments in machine learning and encourages (or discourages, if appropriate) trust in the underlying model. Deep learning approaches can show excellent performance but still have limited practical use if they learn to predict based on confounding factors in a dataset, for instance text labels in the corner of images. By using an explanatory interactive learning approach, with a human expert in the loop during training, it becomes possible to avoid predictions based on confounding factors.

Cyclocarya paliurus (CP) contains triterpene acids that can improve glucose and lipid metabolism disorders. However, controlling the composition and content of these active ingredients in CP extracts is challenging. The main active components in CP triterpene acids, including ursolic acid (UA), oleanolic acid (OA), and betulinic acid (BA), exhibit antihyperglycemic and antihypertensive effects. The response surface methodology was utilized to design and optimize the ratio of UA, OA, and BA based on the inhibition rate of pancrelipase and α-amylase. The proportional mixture of UA, OA, and BA resulted in the formation of a complex known as Cyclocarya paliurus triterpenoid acid (TAC). Se-methylselenocysteine (MSC), a compound with various physiological functions such as antioxidant properties and tumor inhibition, has been used in combination with TAC to form the TAC/MSC complex. Our data demonstrate that TAC/MSC improved palmitic acid (PA)-induced insulin resistance in HepG2 cells through activating the phosphoinositide 3-kinase (PI3K) /protein kinase B (AKT)/glycogen synthase kinase 3 beta (GSK3β) pathway. Moreover, TAC/MSC effectively improved hyperglycemia, glucose intolerance, insulin resistance, and lipid metabolism disorder in mice with type 2 diabetes mellitus (T2DM), attenuated hepatic steatosis, and reduced oxidative stress to alleviate T2DM characteristics.

Cyclocarya paliurus (CP) contains triterpene acids that can improve glucose and lipid metabolism disorders. However, controlling the composition and content of these active ingredients in CP extracts is challenging. The main active components in CP triterpene acids, including ursolic acid (UA), oleanolic acid (OA), and betulinic acid (BA), exhibit antihyperglycemic and antihypertensive effects. The response surface methodology was utilized to design and optimize the ratio of UA, OA, and BA based on the inhibition rate of pancrelipase and α-amylase. The proportional mixture of UA, OA, and BA resulted in the formation of a complex known as Cyclocarya paliurus triterpenoid acid (TAC). Se-methylselenocysteine (MSC), a compound with various physiological functions such as antioxidant properties and tumor inhibition, has been used in combination with TAC to form the TAC/MSC complex. Our data demonstrate that TAC/MSC improved palmitic acid (PA)-induced insulin resistance in HepG2 cells through activating the phosphoinositide 3-kinase (PI3K) /protein kinase B (AKT)/glycogen synthase kinase 3 beta (GSK3β) pathway. Moreover, TAC/MSC effectively improved hyperglycemia, glucose intolerance, insulin resistance, and lipid metabolism disorder in mice with type 2 diabetes mellitus (T2DM), attenuated hepatic steatosis, and reduced oxidative stress to alleviate T2DM characteristics.

Counterfactual examples are an appealing class of post-hoc explanations for machine learning models. Given input \\(x\\) of class \\(y_1\\), its counterfactual is a contrastive example \\(x^\\prime\\) of another class \\(y_0\\). Current approaches primarily solve this task by a complex optimization: define an objective function based on the loss of the counterfactual outcome \\(y_0\\) with hard or soft constraints, then optimize this function as a black-box. This \"deep learning\" approach, however, is rather slow, sometimes tricky, and may result in unrealistic counterfactual examples. In this work, we propose a novel approach to deal with these problems using only two gradient computations based on tractable probabilistic models. First, we compute an unconstrained counterfactual \\(u\\) of \\(x\\) to induce the counterfactual outcome \\(y_0\\). Then, we adapt \\(u\\) to higher density regions, resulting in \\(x^{\\prime}\\). Empirical evidence demonstrates the dominant advantages of our approach.

Deep learning is bringing remarkable contributions to the field of argumentation mining, but the existing approaches still need to fill the gap toward performing advanced reasoning tasks. In this position paper, we posit that neural-symbolic and statistical relational learning could play a crucial role in the integration of symbolic and sub-symbolic methods to achieve this goal.

A key assumption of most statistical machine learning methods is that they have access to independent samples from the distribution of data they encounter at test time. As such, these methods often perform poorly in the face of biased data, which breaks this assumption. In particular, machine learning models have been shown to exhibit Clever-Hans-like behaviour, meaning that spurious correlations in the training set are inadvertently learnt. A number of works have been proposed to revise deep classifiers to learn the right correlations. However, generative models have been overlooked so far. We observe that generative models are also prone to Clever-Hans-like behaviour. To counteract this issue, we propose to debias generative models by disentangling their internal representations, which is achieved via human feedback. Our experiments show that this is effective at removing bias even when human feedback covers only a small fraction of the desired distribution. In addition, we achieve strong disentanglement results in a quantitative comparison with recent methods.

Deep neural networks have shown excellent performances in many real-world applications. Unfortunately, they may show \"Clever Hans\"-like behavior -- making use of confounding factors within datasets -- to achieve high performance. In this work, we introduce the novel learning setting of \"explanatory interactive learning\" (XIL) and illustrate its benefits on a plant phenotyping research task. XIL adds the scientist into the training loop such that she interactively revises the original model via providing feedback on its explanations. Our experimental results demonstrate that XIL can help avoiding Clever Hans moments in machine learning and encourages (or discourages, if appropriate) trust into the underlying model.

Probabilistic graphical models are a central tool in AI; however, they are generally not as expressive as deep neural models, and inference is notoriously hard and slow. In contrast, deep probabilistic models such as sum-product networks (SPNs) capture joint distributions in a tractable fashion, but still lack the expressive power of intractable models based on deep neural networks. Therefore, we introduce conditional SPNs (CSPNs), conditional density estimators for multivariate and potentially hybrid domains which allow harnessing the expressive power of neural networks while still maintaining tractability guarantees. One way to implement CSPNs is to use an existing SPN structure and condition its parameters on the input, e.g., via a deep neural network. This approach, however, might misrepresent the conditional independence structure present in data. Consequently, we also develop a structure-learning approach that derives both the structure and parameters of CSPNs from data. Our experimental evidence demonstrates that CSPNs are competitive with other probabilistic models and yield superior performance on multilabel image classification compared to mean field and mixture density networks. Furthermore, they can successfully be employed as building blocks for structured probabilistic models, such as autoregressive image models.

Nonalcoholic fatty liver disease (NAFLD) affects a large population with incompletely defined mechanism(s). Here we report that Kindlin-2 is dramatically up-regulated in livers in obese mice and patients with NAFLD. Kindlin-2 haploinsufficiency in hepatocytes ameliorates high-fat diet (HFD)-induced NAFLD and glucose intolerance without affecting energy metabolism in mice. In contrast, Kindlin-2 overexpression in liver exacerbates NAFLD and promotes lipid metabolism disorder and inflammation in hepatocytes. A C-terminal region (aa 570-680) of Kindlin-2 binds to and stabilizes Foxo1 by inhibiting its ubiquitination and degradation through the Skp2 E3 ligase. Kindlin-2 deficiency increases Foxo1 phosphorylation at Ser256, which favors its ubiquitination by Skp2. Thus, Kindllin-2 loss down-regulates Foxo1 protein in hepatocytes. Foxo1 overexpression in liver abrogates the ameliorating effect of Kindlin-2 haploinsufficiency on NAFLD in mice. Finally, AAV8-mediated shRNA knockdown of Kindlin-2 in liver alleviates NAFLD in obese mice. Collectively, we demonstrate that Kindlin-2 insufficiency protects against fatty liver by promoting Foxo1 degradation. Here, the authors show that expression of kindlin-2 is increased in patients with nonalcoholic fatty liver disease (NAFLD). In mouse models, specific deletion of kindlin-2 in liver ameliorates, while its overexpression exacerbates, NAFLD by modulating Foxo1 in hepatocytes.

Background and aims Microorganisms play a vital role in regulating soil phosphorus (P) dynamics in terrestrial ecosystems. Here, we investigated the response of soil microbial P cycling potential traits to nitrogen (N) addition via metagenomics and the relationship between microbial potentials and soil P dynamics. Methods Topsoil (0–10 cm) samples were collected from experimental soil that had been maintained for 3 years with low and high level of N addition in an alpine meadow of the Qinghai-Tibet Plateau. Soil microbial functional genes and P fractions were determined. Results The soil available P and microbial biomass P were significantly affected by N inputs and significantly associated with soil properties (including soil pH, alkaline phosphatase activity, and soil total N and NO 3 − -N contents). Meanwhile, high N input decreased the relative abundance of the pstS gene, and low N input reduced the relative abundances of phoB , ugpQ and C-P lyase genes. We further found that the pstS gene was a determinant of soil microbial biomass P and significantly correlated with soil pH. Moreover, Alphaproteobacteria with C-P lyase and Actinobacteria related to alkaline phosphatases and phosphate-specific transport were the most abundant taxa but not affected by N input. Conclusions Short-term high N input could alter soil P dynamics and microbial functional genes. Although there were relationships between the pstS gene, microbial biomass P and soil pH, the microbial functional gene abundance was less important than soil properties in regulating soil P dynamics.