In online advertising, the demand-side platform (a.k.a. DSP) enables advertisers to create different ad creatives for real-time bidding. Intuitively, advertisers tend to create more ad creatives for a single photo to increase the probability of participating in bidding, further enhancing their ad cost. From the perspective of DSP, the following are two overlooked issues. On the one hand, the number of ad creatives cannot grow indefinitely. On the other hand, the marginal effects of ad cost diminish as the number of ad creatives increases. To this end, this paper proposes a two-stage framework named Automated Creatives Quota (ACQ) to achieve the automatic creation and deactivation of ad creatives. ACQ dynamically allocates the creative quota across multiple advertisers to maximize the revenue of the ad platform. ACQ comprises two components: a prediction module to estimate the cost of a photo under different numbers of ad creatives, and an allocation module to decide the quota for photos considering their estimated costs in the prediction module. Specifically, in the prediction module, we develop a multi-task learning model based on an unbalanced binary tree to effectively mitigate the target variable imbalance problem. In the allocation module, we formulate the quota allocation problem as a multiple-choice knapsack problem (MCKP) and develop an efficient solver to solve such large-scale problems involving tens of millions of ads. We performed extensive offline and online experiments to validate the superiority of our proposed framework, which increased cost by 9.34%.

Data-free quantization aims to achieve model quantization without accessing any authentic sample. It is significant in an application-oriented context involving data privacy. Converting noise vectors into synthetic samples through a generator is a popular data-free quantization method, which is called generative data-free quantization. However, there is a difference in attention between synthetic samples and authentic samples. This is always ignored and restricts the quantization performance. First, since synthetic samples of the same class are prone to have homogenous attention, the quantized network can only learn limited modes of attention. Second, synthetic samples in eval mode and training mode exhibit different attention. Hence, the batch-normalization statistics matching tends to be inaccurate. ACQ is proposed in this paper to fix the attention of synthetic samples. An attention center position-condition generator is established regarding the homogenization of intra-class attention. Restricted by the attention center matching loss, the attention center position is treated as the generator's condition input to guide synthetic samples in obtaining diverse attention. Moreover, we design adversarial loss of paired synthetic samples under the same condition to prevent the generator from paying overmuch attention to the condition, which may result in mode collapse. To improve the attention similarity of synthetic samples in different network modes, we introduce a consistency penalty to guarantee accurate BN statistics matching. The experimental results demonstrate that ACQ effectively improves the attention problems of synthetic samples. Under various training settings, ACQ achieves the best quantization performance. For the 4-bit quantization of Resnet18 and Resnet50, ACQ reaches 67.55% and 72.23% accuracy, respectively.

The field of decentralized and distributed learning fits in with the area of modern Internet-of-Things (IoT) and wireless sensor networks (WSN) applications. Due to technological advances and functional advantages related to robustness and scalability [20], decentralized and distributed techniques are becoming more widespread in industry and are the subject of ongoing scientific research. Among various goals specific for learning and inference in decentralized networks, like learning linear modules [15], distributed economic dispatch [6] or target tracking [11], one can point out estimation tasks as in [2] or [4]. In this scenario, sensors (or agents) are scattered around a given area and collect data about an unknown phenomenon, modelled as a nonlinear function m: R R. Due to potential communication restrictions and the lack of dedicated fusion centers, agents may rely only on their local/private measurements and available network information. Following [12] and [7] we begin with brief summary of a few well-known strategies.

Data-free Knowledge Distillation (DFKD) has gained popularity recently, with the fundamental idea of carrying out knowledge transfer from a Teacher neural network to a Student neural network in the absence of training data. However, in the Adversarial DFKD framework, the student network's accuracy, suffers due to the non-stationary distribution of the pseudo-samples under multiple generator updates. To this end, at every generator update, we aim to maintain the student's performance on previously encountered examples while acquiring knowledge from samples of the current distribution. Thus, we propose a meta-learning inspired framework by treating the task of Knowledge-Acquisition (learning from newly generated samples) and Knowledge-Retention (retaining knowledge on previously met samples) as meta-train and meta-test, respectively. Hence, we dub our method as Learning to Retain while Acquiring. Moreover, we identify an implicit aligning factor between the Knowledge-Retention and Knowledge-Acquisition tasks indicating that the proposed student update strategy enforces a common gradient direction for both tasks, alleviating interference between the two objectives. Finally, we support our hypothesis by exhibiting extensive evaluation and comparison of our method with prior arts on multiple datasets.

Recently proposed methods in data subset selection, that is active learning and active sampling, use Fisher information, Hessians, similarity matrices based on gradients, and gradient lengths to estimate how informative data is for a model's training. Are these different approaches connected, and if so, how? We revisit the fundamentals of Bayesian optimal experiment design and show that these recently proposed methods can be understood as approximations to information-theoretic quantities: among them, the mutual information between predictions and model parameters, known as expected information gain or BALD in machine learning, and the mutual information between predictions of acquisition candidates and test samples, known as expected predictive information gain. We develop a comprehensive set of approximations using Fisher information and observed information and derive a unified framework that connects seemingly disparate literature. Although Bayesian methods are often seen as separate from non-Bayesian ones, the sometimes fuzzy notion of "informativeness" expressed in various non-Bayesian objectives leads to the same couple of information quantities, which were, in principle, already known by Lindley (1956) and MacKay (1992).

In active learning, new labels are commonly acquired in batches. However, common acquisition functions are only meant for one-sample acquisition rounds at a time, and when their scores are used naively for batch acquisition, they result in batches lacking diversity, which deteriorates performance. On the other hand, state-of-the-art batch acquisition functions are costly to compute. In this paper, we present a novel class of stochastic acquisition functions that extend one-sample acquisition functions to the batch setting by observing how one-sample acquisition scores change as additional samples are acquired and modelling this difference for additional batch samples. We simply acquire new samples by sampling from the pool set using a Gibbs distribution based on the acquisition scores. Our acquisition functions are both vastly cheaper to compute and out-perform other batch acquisition functions.

Online shopping has accelerated with a notable 30% increase in digital sales in the U.S.; this means that previously nascent or hard-to-sell-in technologies are now being considered by major retailers. Artificial intelligence is one of these technologies that is seeing more rapid integration, with $7.3 billion projected to be spent by retailers in 2022. And, last month's Google release of new AI-powered e-commerce tools highlights how far down the channel AI adoption has become. As a serial tech entrepreneur currently operating in the next-gen AI space, I'm seeing that AI is no longer relegated to product recommendation engines but is quickly becoming the backbone of retail. The pandemic exposed failures within third-party (3P) warehousing and last-mile delivery operations, many of these related to shipping errors, complex deliveries and faulty projections on timing.

For the past few years, artificial intelligence has largely been used for marketing analytics. However, as we push toward a more democratized AI future, AI will become a valuable tool in the arsenal of all marketers. In this instance, democratization means that employees at all levels of an organization will feel empowered to make final decisions guided by artificial intelligence and then act on those decisions. In short, the democratization will be a flattening of judgment-based decision making. For marketers, this will mean more critical analysis to support everyday business decisions.

As I watch 2021 approach, I have both trepidation and hope. I have concern over the multitude of ways in which the world will change around me, but I also have hope for the promising opportunities in our future. The more challenges we have, the more chances we have to build a world that better suits our needs. I do believe that great technology underpins those new solutions, and I'm more and more hopeful about the ways in which AI, like the emergent technologies before it, can spark long-lasting and impactful change. In 2021, I predict we will see two main evolutions in artificial intelligence: better business integration and more democratization.

One of the main applications of description logics is the ontology-based data access model, which requires algorithms for query answering over ontologies. In fact, some description logics, like those in the DL-Lite family, are designed so that simple queries, such as conjunctive queries, are efficiently computable. In this paper we study counting aggregate queries over ontologies, i.e. queries which use aggregate functions COUNT and COUNT DISTINCT. We propose an intuitive semantics for certain answers for these queries, which conforms to the open world assumption. We compare our semantics with other approaches that have been proposed in different contexts. We establish data and combined computational complexity for the problems of answering counting aggregate queries over ontologies for several variants of DL-Lite.