Extracting spatial-temporal knowledge from data is useful in many applications. It is important that the obtained knowledge is human-interpretable and amenable to formal analysis. In this paper, we propose a method that trains neural networks to learn spatial-temporal properties in the form of weighted graph-based signal temporal logic (wGSTL) formulas. For learning wGSTL formulas, we introduce a flexible wGSTL formula structure in which the user's preference can be applied in the inferred wGSTL formulas. In the proposed framework, each neuron of the neural networks corresponds to a subformula in a flexible wGSTL formula structure. We initially train a neural network to learn the wGSTL operators and then train a second neural network to learn the parameters in a flexible wGSTL formula structure. We use a COVID-19 dataset and a rain prediction dataset to evaluate the performance of the proposed framework and algorithms. We compare the performance of the proposed framework with three baseline classification methods including K-nearest neighbors, decision trees, and artificial neural networks. The classification accuracy obtained by the proposed framework is comparable with the baseline classification methods.

Contrastive explanations for understanding the behavior of black box models has gained a lot of attention recently as they provide potential for recourse. In this paper, we propose a method Contrastive Attributed explanations for Text (CAT) which provides contrastive explanations for natural language text data with a novel twist as we build and exploit attribute classifiers leading to more semantically meaningful explanations. To ensure that our contrastive generated text has the fewest possible edits with respect to the original text, while also being fluent and close to a human generated contrastive, we resort to a minimal perturbation approach regularized using a BERT language model and attribute classifiers trained on available attributes. We show through qualitative examples and a user study that our method not only conveys more insight because of these attributes, but also leads to better quality (contrastive) text. Moreover, quantitatively we show that our method is more efficient than other state-of-the-art methods with it also scoring higher on benchmark metrics such as flip rate, (normalized) Levenstein distance, fluency and content preservation.

Estimating the performance of a machine learning system is a longstanding challenge in artificial intelligence research. Today, this challenge is especially relevant given the emergence of systems which appear to increasingly outperform human beings. In some cases, this "superhuman" performance is readily demonstrated; for example by defeating legendary human players in traditional two player games. On the other hand, it can be challenging to evaluate classification models that potentially surpass human performance. Indeed, human annotations are often treated as a ground truth, which implicitly assumes the superiority of the human over any models trained on human annotations. In reality, human annotators can make mistakes and be subjective. Evaluating the performance with respect to a genuine oracle may be more objective and reliable, even when querying the oracle is expensive or impossible. In this paper, we first raise the challenge of evaluating the performance of both humans and models with respect to an oracle which is unobserved. We develop a theory for estimating the accuracy compared to the oracle, using only imperfect human annotations for reference. Our analysis provides a simple recipe for detecting and certifying superhuman performance in this setting, which we believe will assist in understanding the stage of current research on classification. We validate the convergence of the bounds and the assumptions of our theory on carefully designed toy experiments with known oracles. Moreover, we demonstrate the utility of our theory by meta-analyzing large-scale natural language processing tasks, for which an oracle does not exist, and show that under our assumptions a number of models from recent years are with high probability superhuman.

Discovering relevant patterns for a particular user remains a challenging task in data mining. In real-life applications, relevant patterns may be either frequent or rare ones in the data. In itemset mining, setting the minimum support threshold is a real dilemma (a high value misses rare itemsets, a low value generates a large number of meaningless itemsets). To tackle the rare item problem [8], several approaches were proposed to mine frequent pattern with multiple minimum supports. In [8], the problem of mining frequent itemsets with multiple Minimum Item Supports (MIS) was introduced with a first revision of Apriori algorithm (MSApriori). Then, other Apriori-like approaches were proposed like MMS Cumulate and MMS Stratify [12]. The well-known FPGrowth was extended with a condensed FP-tree structure to mine frequent itemsets with multiple MIS (CFPGrowth [5], CFPGrowth [6]). In [3], FP ME was proposed based on set-enumeration-tree structure and sorted downward closure property. The specialized algorithms introduced previously are effective for mining patterns with multiple MIS.

Grounded dialogue models generate responses that are grounded on certain concepts. Limited by the distribution of grounded dialogue data, models trained on such data face the transferability challenges in terms of the data distribution and the type of grounded concepts. To address the challenges, we propose the grounded minimal editing framework, which minimally edits existing responses to be grounded on the given concept. Focusing on personas, we propose Grounded Minimal Editor (GME), which learns to edit by disentangling and recombining persona-related and persona-agnostic parts of the response. To evaluate persona-grounded minimal editing, we present the PersonaMinEdit dataset, and experimental results show that GME outperforms competitive baselines by a large margin. To evaluate the transferability, we experiment on the test set of BlendedSkillTalk and show that GME can edit dialogue models' responses to largely improve their persona consistency while preserving the use of knowledge and empathy.

The U.N. human rights chief is calling for a moratorium on the use of artificial intelligence technology that poses a serious risk to human rights, including face-scanning systems that track people in public spaces. Michelle Bachelet, the U.N. High Commissioner for Human Rights, also said Wednesday that countries should expressly ban AI applications which don't comply with international human rights law. Applications that should be prohibited include government "social scoring" systems that judge people based on their behavior and certain AI-based tools that categorize people into clusters such as by ethnicity or gender. AI-based technologies can be a force for good but they can also "have negative, even catastrophic, effects if they are used without sufficient regard to how they affect people's human rights," Bachelet said in a statement. Her comments came along with a new U.N. report that examines how countries and businesses have rushed into applying AI systems that affect people's lives and livelihoods without setting up proper safeguards to prevent discrimination and other harms.

Metigy, the world's leading artificial intelligence (AI) digital marketing solution for SMEs, has partnered with Social Status, a provider of social media analytics and reporting tools, to understand the impact of AI-powered marketing strategies. Point B Establishes Partnership with Vettd.ai, Together, Metigy and Social Status explored the metrics that matter when measuring the success of a digital marketing campaign to produce the Social Media Benchmarking Report: Q2 2021. The new report analysed hundreds of thousands of social media posts across 11,000 Social Status accounts and compared the findings to 5,000 Metigy accounts during Q2 2021. It revealed that savvy businesses using AI as part of their social media strategies saw greater results than the benchmark standards.

The Transform Technology Summits start October 13th with Low-Code/No Code: Enabling Enterprise Agility. Constructor, a San Francisco, California-based ecommerce personalization startup, today announced that it raised $55 million in a series A round led by Silversmith Capital Partners. The funding, which brings the company's total raised to $61.5 million, will be put toward product development, hiring, and go-to-market efforts, according to CEO Eli Finkelshteyn. Online commerce conversions are well behind in-store -- the average online shop sees less than 3% in conversions. But even though $4.2 trillion was spent on ecommerce platforms in 2020 alone, few ecommerce retailers have invested in a digital merchandising strategy.

Editor's note: Sourav Mazumder is a speaker for ODSC West 2021. Be sure to check out his talk, "Operationalization of Models Developed and Deployed in Heterogeneous Platforms," for more info on trustworthy AI there. Artificial intelligence (AI) is already having a significant impact on the development of humanity, already. For enterprises, the use of AI is not an option anymore. However, the core of AI relies on the use of data samples/examples to train a system/machine using algorithms so that it can behave intelligently like a human.

Our journey with AWS Glue was a bit of a struggle once we started to dig deeper into the streaming functionality of it, the orchestration of so many layers added a huge overhead that we weren't expecting and whilst most of that is handled within the AWS suite of products, there are just too many benefits to switching our pipelines over to GCP and BigQuery to be ignored. Next steps are to finalise our deployment by using Cloud Composer (Airflow) to orchestrate the creation of each of the tables and provide a monitoring dashboard to help us detect failures and act on them. I will say that AWS got in touch with me after my previous article and I got on a call with the AWS Glue product team, in their words I had "hit pretty much every sharp edge possible" (seems to be a running theme with me -- perhaps I should switch careers to QA engineer?),