Despite abundant negotiation strategies in literature, the complexity of automated negotiation forbids a single strategy from being dominant against all others in different negotiation scenarios. To overcome this, one approach is to use mixture of experts, but at the same time, one problem of this method is the selection of experts, as this approach is limited by the competency of the experts selected. Another problem with most negotiation strategies is their incapability of adapting to dynamic variation of the opponent's behaviour within a single negotiation session resulting in poor performance. This work focuses on both, solving the problem of expert selection and adapting to the opponent's behaviour with our Autonomous Negotiating Agent Framework. This framework allows real-time classification of opponent's behaviour and provides a mechanism to select, switch or combine strategies within a single negotiation session. Additionally, our framework has a reviewer component which enables self-enhancement capability by deciding to include new strategies or replace old ones with better strategies periodically. We demonstrate an instance of our framework by implementing maximum entropy reinforcement learning based strategies with a deep learning based opponent classifier. Finally, we evaluate the performance of our agent against state-of-the-art negotiators under varied negotiation scenarios.

In this paper, we study the problem of learning to satisfy temporal logic specifications with a group of agents in an unknown environment, which may exhibit probabilistic behaviour. From a learning perspective these specifications provide a rich formal language with which to capture tasks or objectives, while from a logic and automated verification perspective the introduction of learning capabilities allows for practical applications in large, stochastic, unknown environments. The existing work in this area is, however, limited. Of the frameworks that consider full linear temporal logic or have correctness guarantees, all methods thus far consider only the case of a single temporal logic specification and a single agent. In order to overcome this limitation, we develop the first multi-agent reinforcement learning technique for temporal logic specifications, which is also novel in its ability to handle multiple specifications. We provide correctness and convergence guarantees for our main algorithm - ALMANAC (Automaton/Logic Multi-Agent Natural Actor-Critic) - even when using function approximation. Alongside our theoretical results, we further demonstrate the applicability of our technique via a set of preliminary experiments.

Post-harvest diseases of apple are one of the major issues in the economical sector of apple production, causing severe economical losses to producers. Thus, we developed DSSApple, a picture-based decision support system able to help users in the diagnosis of apple diseases. Specifically, this paper addresses the problem of sequentially optimizing for the best diagnosis, leveraging past interactions with the system and their contextual information (i.e. the evidence provided by the users). The problem of learning an online model while optimizing for its outcome is commonly addressed in the literature through a stochastic active learning paradigm - i.e. Contextual Multi-Armed Bandit (CMAB). This methodology interactively updates the decision model considering the success of each past interaction with respect to the context provided in each round. However, this information is very often partial and inadequate to handle such complex decision making problems. On the other hand, human decisions implicitly include unobserved factors (referred in the literature as unobserved confounders) that significantly contribute to the human's final decision. In this paper, we take advantage of the information embedded in the observed human decisions to marginalize confounding factors and improve the capability of the CMAB model to identify the correct diagnosis. Specifically, we propose a Counterfactual Contextual Multi-Armed Bandit, a model based on the causal concept of counterfactual. The proposed model is validated with offline experiments based on data collected through a large user study on the application. The results prove that our model is able to outperform both traditional CMAB algorithms and observed user decisions, in real-world tasks of predicting the correct apple disease.

Although 3D Convolutional Neural Networks are essential for most learning based applications involving dense 3D data, their applicability is limited due to excessive memory and computational requirements. Compressing such networks by pruning therefore becomes highly desirable. However, pruning 3D CNNs is largely unexplored possibly because of the complex nature of typical pruning algorithms that embeds pruning into an iterative optimization paradigm. In this work, we introduce a Resource Aware Neuron Pruning (RANP) algorithm that prunes 3D CNNs at initialization to high sparsity levels. Specifically, the core idea is to obtain an importance score for each neuron based on their sensitivity to the loss function. This neuron importance is then reweighted according to the neuron resource consumption related to FLOPs or memory. We demonstrate the effectiveness of our pruning method on 3D semantic segmentation with widely used 3D-UNets on ShapeNet and BraTS'18 datasets, video classification with MobileNetV2 and I3D on UCF101 dataset, and two-view stereo matching with Pyramid Stereo Matching (PSM) network on SceneFlow dataset. In these experiments, our RANP leads to roughly 50%-95% reduction in FLOPs and 35%-80% reduction in memory with negligible loss in accuracy compared to the unpruned networks. This significantly reduces the computational resources required to train 3D CNNs. The pruned network obtained by our algorithm can also be easily scaled up and transferred to another dataset for training.

Recently a number of empirical "universal" scaling law papers have been published, most notably by OpenAI. `Scaling laws' refers to power-law decreases of training or test error w.r.t. more data, larger neural networks, and/or more compute. In this work we focus on scaling w.r.t. data size $n$. Theoretical understanding of this phenomenon is largely lacking, except in finite-dimensional models for which error typically decreases with $n^{-1/2}$ or $n^{-1}$, where $n$ is the sample size. We develop and theoretically analyse the simplest possible (toy) model that can exhibit $n^{-\beta}$ learning curves for arbitrary power $\beta>0$, and determine whether power laws are universal or depend on the data distribution.

Recent research shows -- somewhat astonishingly -- that people are willing to ascribe moral blame to AI-driven systems when they cause harm [1]-[4]. In this paper, we explore the moral-psychological underpinnings of these findings. Our hypothesis was that the reason why people ascribe moral blame to AI systems is that they consider them capable of entertaining inculpating mental states (what is called mens rea in the law). To explore this hypothesis, we created a scenario in which an AI system runs a risk of poisoning people by using a novel type of fertilizer. Manipulating the computational (or quasi-cognitive) abilities of the AI system in a between-subjects design, we tested whether people's willingness to ascribe knowledge of a substantial risk of harm (i.e., recklessness) and blame to the AI system. Furthermore, we investigated whether the ascription of recklessness and blame to the AI system would influence the perceived blameworthiness of the system's user (or owner). In an experiment with 347 participants, we found (i) that people are willing to ascribe blame to AI systems in contexts of recklessness, (ii) that blame ascriptions depend strongly on the willingness to attribute recklessness and (iii) that the latter, in turn, depends on the perceived "cognitive" capacities of the system. Furthermore, our results suggest (iv) that the higher the computational sophistication of the AI system, the more blame is shifted from the human user to the AI system.

The capabilities of natural language models trained on large-scale data have increased immensely over the past few years. Downstream applications are at risk of inheriting biases contained in these models, with potential negative consequences especially for marginalized groups. In this paper, we analyze the occupational biases of a popular generative language model, GPT-2, intersecting gender with five protected categories: religion, sexuality, ethnicity, political affiliation, and name origin. Using a novel data collection pipeline we collect 396k sentence completions of GPT-2 and find: (i) The machine-predicted jobs are less diverse and more stereotypical for women than for men, especially for intersections; (ii) Fitting 262 logistic models shows intersectional interactions to be highly relevant for occupational associations; (iii) For a given job, GPT-2 reflects the societal skew of gender and ethnicity in the US, and in some cases, pulls the distribution towards gender parity, raising the normative question of what language models _should_ learn.

Deep graph embedding is an important approach for community discovery. Deep graph neural network with self-supervised mechanism can obtain the low-dimensional embedding vectors of nodes from unlabeled and unstructured graph data. The high-order information of graph can provide more abundant structure information for the representation learning of nodes. However, most self-supervised graph neural networks only use adjacency matrix as the input topology information of graph and cannot obtain too high-order information since the number of layers of graph neural network is fairly limited. If there are too many layers, the phenomenon of over smoothing will appear. Therefore how to obtain and fuse high-order information of graph by a shallow graph neural network is an important problem. In this paper, a deep graph embedding algorithm with self-supervised mechanism for community discovery is proposed. The proposed algorithm uses self-supervised mechanism and different high-order information of graph to train multiple deep graph convolution neural networks. The outputs of multiple graph convolution neural networks are fused to extract the representations of nodes which include the attribute and structure information of a graph. In addition, data augmentation and negative sampling are introduced into the training process to facilitate the improvement of embedding result. The proposed algorithm and the comparison algorithms are conducted on the five experimental data sets. The experimental results show that the proposed algorithm outperforms the comparison algorithms on the most experimental data sets. The experimental results demonstrate that the proposed algorithm is an effective algorithm for community discovery.

Human settlement has always been dependent on a stable supply of clean water nearby. With the increase in global population and a decline in the quality of our freshwater resources, we are constantly looking for technologies that will ensure a reliable supply of clean water. The Union Budget 2021-22 announced Jal Jeevan Mission (Urban) to bring safe water to 2.86 Cr households through tap connection. This in line with the Centre's rural water supply project launched in 2019. Finance minister Nirmala Sitharaman announced an outlay of INR 50,011 Cr for this scheme.

In this paper, we study the generalization properties of Model-Agnostic Meta-Learning (MAML) algorithms for supervised learning problems. We focus on the setting in which we train the MAML model over $m$ tasks, each with $n$ data points, and characterize its generalization error from two points of view: First, we assume the new task at test time is one of the training tasks, and we show that, for strongly convex objective functions, the expected excess population loss is bounded by $\mathcal{O}(1/mn)$. Second, we consider the MAML algorithm's generalization to an unseen task and show that the resulting generalization error depends on the total variation distance between the underlying distributions of the new task and the tasks observed during the training process. Our proof techniques rely on the connections between algorithmic stability and generalization bounds of algorithms. In particular, we propose a new definition of stability for meta-learning algorithms, which allows us to capture the role of both the number of tasks $m$ and number of samples per task $n$ on the generalization error of MAML.