Chengrun Yang, Jicong Fan, Ziyang Wu, and Madeleine Udell This is an extended version of AutoML Pipeline Selection: Efficiently Navigating the Combinatorial Space (DOI: 10.1145/3394486.3403197) at the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2020. Abstract--Data scientists seeking a good supervised learning model on a new dataset have many choices to make: they must preprocess the data, select features, possibly reduce the dimension, select an estimation algorithm, and choose hyperparameters for each of these pipeline components. With new pipeline components comes a combinatorial explosion in the number of choices! In this work, we design a new AutoML system to address this challenge: an automated system to design a supervised learning pipeline. Our system uses matrix and tensor factorization as surrogate models to model the combinatorial pipeline search space.

Human feedback is widely used to train agents in many domains. However, previous works rarely consider the uncertainty when humans provide feedback, especially in cases that the optimal actions are not obvious to the trainers. For example, the reward of a sub-optimal action can be stochastic and sometimes exceeds that of the optimal action, which is common in games or real-world. Trainers are likely to provide positive feedback to sub-optimal actions, negative feedback to the optimal actions and even do not provide feedback in some confusing situations. Existing works, which utilize the Expectation Maximization (EM) algorithm and treat the feedback model as hidden parameters, do not consider uncertainties in the learning environment and human feedback. To address this challenge, we introduce a novel feedback model that considers the uncertainty of human feedback. However, this incurs intractable calculus in the EM algorithm. To this end, we propose a novel approximate EM algorithm, in which we approximate the expectation step with the Gradient Descent method. Experimental results in both synthetic scenarios and two real-world scenarios with human participants demonstrate the superior performance of our proposed approach.

Active inference offers a first principle account of sentient behaviour, from which special and important cases can be derived, e.g., reinforcement learning, active learning, Bayes optimal inference, Bayes optimal design, etc. Active inference resolves the exploitation-exploration dilemma in relation to prior preferences, by placing information gain on the same footing as reward or value. In brief, active inference replaces value functions with functionals of (Bayesian) beliefs, in the form of an expected (variational) free energy. In this paper, we consider a sophisticated kind of active inference, using a recursive form of expected free energy. Sophistication describes the degree to which an agent has beliefs about beliefs. We consider agents with beliefs about the counterfactual consequences of action for states of affairs and beliefs about those latent states. In other words, we move from simply considering beliefs about "what would happen if I did that" to "what would I believe about what would happen if I did that". The recursive form of the free energy functional effectively implements a deep tree search over actions and outcomes in the future. Crucially, this search is over sequences of belief states, as opposed to states per se. We illustrate the competence of this scheme, using numerical simulations of deep decision problems.

Recent work in the behavioural sciences has begun to overturn the long-held belief that human decision making is irrational, suboptimal and subject to biases. This turn to the rational suggests that human decision making may be a better source of ideas for constraining how machine learning problems are defined than would otherwise be the case. One promising idea concerns human decision making that is dependent on apparently irrelevant aspects of the choice context. Previous work has shown that by taking into account choice context and making relational observations, people can maximize expected value. Other work has shown that Partially observable Markov decision processes (POMDPs) are a useful way to formulate human-like decision problems. Here, we propose a novel POMDP model for contextual choice tasks and show that, despite the apparent irrationalities, a reinforcement learner can take advantage of the way that humans make decisions. We suggest that human irrationalities may offer a productive source of inspiration for improving the design of AI architectures and machine learning methods.

Much of human learning and inference can be framed within the computational problem of relational generalization. In this project, we propose a Bayesian model that generalizes relational knowledge to novel environments by analogically weighting predictions from previously encountered relational structures. First, we show that this learner outperforms a naive, theory-based learner on relational data derived from random- and Wikipedia-based systems when experience with the environment is small. Next, we show how our formalization of analogical similarity translates to the selection and weighting of analogies. Finally, we combine the analogy- and theory-based learners in a single nonparametric Bayesian model, and show that optimal relational generalization transitions from relying on analogies to building a theory of the novel system with increasing experience in it. Beyond predicting unobserved interactions better than either baseline, this formalization gives a computational-level perspective on the formation and abstraction of analogies themselves.

Active inference is a Bayesian framework for understanding biological intelligence. The underlying theory brings together perception and action under one single imperative: minimizing free energy. However, despite its theoretical utility in explaining intelligence, computational implementations have been restricted to low-dimensional and idealized situations. In this paper, we present a neural architecture for building deep active inference agents operating in complex, continuous state-spaces using multiple forms of Monte-Carlo (MC) sampling. For this, we introduce a number of techniques, novel to active inference. These include: i) selecting free-energy-optimal policies via MC tree search, ii) approximating this optimal policy distribution via a feed-forward `habitual' network, iii) predicting future parameter belief updates using MC dropouts and, finally, iv) optimizing state transition precision (a high-end form of attention). Our approach enables agents to learn environmental dynamics efficiently, while maintaining task performance, in relation to reward-based counterparts. We illustrate this in a new toy environment, based on the dSprites data-set, and demonstrate that active inference agents automatically create disentangled representations that are apt for modeling state transitions. In a more complex Animal-AI environment, our agents (using the same neural architecture) are able to simulate future state transitions and actions (i.e., plan), to evince reward-directed navigation - despite temporary suspension of visual input. These results show that deep active inference - equipped with MC methods - provides a flexible framework to develop biologically-inspired intelligent agents, with applications in both machine learning and cognitive science.

Real world multi-agent tasks often involve varying types and quantities of agents and non-agent entities. Agents frequently do not know a priori how many other agents and non-agent entities they will need to interact with in order to complete a given task, requiring agents to generalize across a combinatorial number of task configurations with each potentially requiring different strategies. In this work, we tackle the problem of multi-agent reinforcement learning (MARL) in such dynamic scenarios. We hypothesize that, while the optimal behaviors in these scenarios with varying quantities and types of agents/entities are diverse, they may share common patterns within sub-teams of agents that are combined to form team behavior. As such, we propose a method that can learn these subgroup relationships and how they can be combined, ultimately improving knowledge sharing and generalization across scenarios. This method, Attentive-Imaginative QMIX, extends QMIX for dynamic MARL in two ways: 1) an attention mechanism that enables model sharing across variable sized scenarios and 2) a training objective that improves learning across scenarios with varying combinations of agent/entity types by factoring the value function into imagined sub-scenarios. We validate our approach on both a novel grid-world task as well as a version of the StarCraft Multi-Agent Challenge [28] minimally modified for the dynamic scenario setting.

Deep neural networks are often ignorant about what they do not know and overconfident when they make uninformed predictions. Some recent approaches quantify classification uncertainty directly by training the model to output high uncertainty for the data samples close to class boundaries or from the outside of the training distribution. These approaches use an auxiliary data set during training to represent out-of-distribution samples. However, selection or creation of such an auxiliary data set is non-trivial, especially for high dimensional data such as images. In this work we develop a novel neural network model that is able to express both aleatoric and epistemic uncertainty to distinguish decision boundary and out-of-distribution regions of the feature space. To this end, variational autoencoders and generative adversarial networks are incorporated to automatically generate out-of-distribution exemplars for training. Through extensive analysis, we demonstrate that the proposed approach provides better estimates of uncertainty for in- and out-of-distribution samples, and adversarial examples on well-known data sets against state-of-the-art approaches including recent Bayesian approaches for neural networks and anomaly detection methods.

We consider the contextual bandit problem, where a player sequentially makes decisions based on past observations to maximize the cumulative reward. Although many algorithms have been proposed for contextual bandit, most of them rely on finding the maximum likelihood estimator at each iteration, which requires $O(t)$ time at the $t$-th iteration and are memory inefficient. A natural way to resolve this problem is to apply online stochastic gradient descent (SGD) so that the per-step time and memory complexity can be reduced to constant with respect to $t$, but a contextual bandit policy based on online SGD updates that balances exploration and exploitation has remained elusive. In this work, we show that online SGD can be applied to the generalized linear bandit problem. The proposed SGD-TS algorithm, which uses a single-step SGD update to exploit past information and uses Thompson Sampling for exploration, achieves $\tilde{O}(\sqrt{dT})$ regret with the total time complexity that scales linearly in $T$ and $d$, where $T$ is the total number of rounds and $d$ is the number of features. Experimental results show that SGD-TS consistently outperforms existing algorithms on both synthetic and real datasets.

We present a method for automatically identifying the causal structure of a dynamical control system. Through a suitable experiment design and subsequent causal analysis, the method reveals, which state and input variables of the system have a causal influence on each other. The experiment design builds on the concept of controllability, which provides a systematic way to compute input trajectories that steer the system to specific regions in its state space. For the causal analysis, we leverage powerful techniques from causal inference and extend them to control systems. Further, we derive conditions that guarantee discovery of the true causal structure of the system and show that the obtained knowledge of the causal structure reduces the complexity of model learning and yields improved generalization capabilities. Experiments on a robot arm demonstrate reliable causal identification from real-world data and extrapolation to regions outside the training domain.