Tuning hyperparameters is a crucial but arduous part of the machine learning pipeline. Hyperparameter optimization is even more challenging in federated learning, where models are learned over a distributed network of heterogeneous devices; here, the need to keep data on device and perform local training makes it difficult to efficiently train and evaluate configurations. In this work, we investigate the problem of federated hyperparameter tuning. We first identify key challenges and show how standard approaches may be adapted to form baselines for the federated setting. Then, by making a novel connection to the neural architecture search technique of weight-sharing, we introduce a new method, FedEx, to accelerate federated hyperparameter tuning that is applicable to widely-used federated optimization methods such as FedAvg and recent variants. Theoretically, we show that a FedEx variant correctly tunes the on-device learning rate in the setting of online convex optimization across devices. Empirically, we show that FedEx can outperform natural baselines for federated hyperparameter tuning by several percentage points on the Shakespeare, FEMNIST, and CIFAR-10 benchmarks--obtaining higher accuracy using the same training budget.

Natural policy gradient (NPG) methods with function approximation achieve impressive empirical success in reinforcement learning problems with large state-action spaces. However, theoretical understanding of their convergence behaviors remains limited in the function approximation setting. In this paper, we perform a finite-time analysis of NPG with linear function approximation and softmax parameterization, and prove for the first time that widely used entropy regularization method, which encourages exploration, leads to linear convergence rate. We adopt a Lyapunov drift analysis to prove the convergence results and explain the effectiveness of entropy regularization in improving the convergence rates.

We present the Multiple Criteria Decision Analysis Methods Selection Software (MCDA-MSS). This decision support system helps analysts answering a recurring question in decision science: Which is the most suitable Multiple Criteria Decision Analysis method (or a subset of MCDA methods) that should be used for a given Decision-Making Problem (DMP)?. The MCDA-MSS includes guidance to lead decision-making processes and choose among an extensive collection (over 200) of MCDA methods. These are assessed according to an original comprehensive set of problem characteristics. The accounted features concern problem formulation, preference elicitation and types of preference information, desired features of a preference model, and construction of the decision recommendation. The applicability of the MCDA-MSS has been tested on several case studies. The MCDA-MSS includes the capabilities of (i) covering from very simple to very complex DMPs, (ii) offering recommendations for DMPs that do not match any method from the collection, (iii) helping analysts prioritize efforts for reducing gaps in the description of the DMPs, and (iv) unveiling methodological mistakes that occur in the selection of the methods. A community-wide initiative involving experts in MCDA methodology, analysts using these methods, and decision-makers receiving decision recommendations will contribute to expansion of the MCDA-MSS.

We consider the problem of teaching via demonstrations in sequential decision-making settings. In particular, we study how to design a personalized curriculum over demonstrations to speed up the learner's convergence. We provide a unified curriculum strategy for two popular learner models: Maximum Causal Entropy Inverse Reinforcement Learning (MaxEnt-IRL) and Cross-Entropy Behavioral Cloning (CrossEnt-BC). Our unified strategy induces a ranking over demonstrations based on a notion of difficulty scores computed w.r.t. the teacher's optimal policy and the learner's current policy. Compared to the state of the art, our strategy doesn't require access to the learner's internal dynamics and still enjoys similar convergence guarantees under mild technical conditions. Furthermore, we adapt our curriculum strategy to teach a learner using domain knowledge in the form of task-specific difficulty scores when the teacher's optimal policy is unknown. Experiments on a car driving simulator environment and shortest path problems in a grid-world environment demonstrate the effectiveness of our proposed curriculum strategy.

We present Meta Learning for Knowledge Distillation (MetaDistil), a simple yet effective alternative to traditional knowledge distillation (KD) methods where the teacher model is fixed during training. We show the teacher network can learn to better transfer knowledge to the student network (i.e., learning to teach) with the feedback from the performance of the distilled student network in a meta learning framework. Moreover, we introduce a pilot update mechanism to improve the alignment between the inner-learner and meta-learner in meta learning algorithms that focus on an improved inner-learner. Experiments on various benchmarks show that MetaDistil can yield significant improvements compared with traditional KD algorithms and is less sensitive to the choice of different student capacity and hyperparameters, facilitating the use of KD on different tasks and models. With the prevalence of large neural networks with millions or billions of parameters, model compression is gaining prominence for facilitating efficient, eco-friendly deployment for machine learning applications. Previous works often train a large model as the "teacher"; then they fix the teacher and train a "student" model to mimic the behavior of the teacher, in order to transfer the knowledge from the teacher to the student. However, this paradigm has the following drawbacks: (1) The teacher is unaware of the student. Recent studies in pedagogy suggest student-centered learning, which considers students' characteristics and learning capability, has shown effectiveness improving students' performance (Cornelius-White, 2007; Wright, 2011).

Biological agents have meaningful interactions with their environment despite the absence of a reward signal. In such instances, the agent can learn preferred modes of behaviour that lead to predictable states -- necessary for survival. In this paper, we pursue the notion that this learnt behaviour can be a consequence of reward-free preference learning that ensures an appropriate trade-off between exploration and preference satisfaction. For this, we introduce a model-based Bayesian agent equipped with a preference learning mechanism (pepper) using conjugate priors. These conjugate priors are used to augment the expected free energy planner for learning preferences over states (or outcomes) across time. Importantly, our approach enables the agent to learn preferences that encourage adaptive behaviour at test time. We illustrate this in the OpenAI Gym FrozenLake and the 3D mini-world environments -- with and without volatility. Given a constant environment, these agents learn confident (i.e., precise) preferences and act to satisfy them. Conversely, in a volatile setting, perpetual preference uncertainty maintains exploratory behaviour. Our experiments suggest that learnable (reward-free) preferences entail a trade-off between exploration and preference satisfaction. Pepper offers a straightforward framework suitable for designing adaptive agents when reward functions cannot be predefined as in real environments.

Intelligent and adaptive online education systems aim to make high-quality education available for a diverse range of students. However, existing systems usually depend on a pool of hand-made questions, limiting how fine-grained and open-ended they can be in adapting to individual students. We explore targeted question generation as a controllable sequence generation task. We first show how to fine-tune pre-trained language models for deep knowledge tracing (LM-KT). This model accurately predicts the probability of a student answering a question correctly, and generalizes to questions not seen in training. We then use LM-KT to specify the objective and data for training a model to generate questions conditioned on the student and target difficulty. Our results show we succeed at generating novel, well-calibrated language translation questions for second language learners from a real online education platform.

We present DistillFlow, a knowledge distillation approach to learning optical flow. DistillFlow trains multiple teacher models and a student model, where challenging transformations are applied to the input of the student model to generate hallucinated occlusions as well as less confident predictions. Then, a self-supervised learning framework is constructed: confident predictions from teacher models are served as annotations to guide the student model to learn optical flow for those less confident predictions. The self-supervised learning framework enables us to effectively learn optical flow from unlabeled data, not only for non-occluded pixels, but also for occluded pixels. DistillFlow achieves state-of-the-art unsupervised learning performance on both KITTI and Sintel datasets. Our self-supervised pre-trained model also provides an excellent initialization for supervised fine-tuning, suggesting an alternate training paradigm in contrast to current supervised learning methods that highly rely on pre-training on synthetic data. At the time of writing, our fine-tuned models ranked 1st among all monocular methods on the KITTI 2015 benchmark, and outperform all published methods on the Sintel Final benchmark. More importantly, we demonstrate the generalization capability of DistillFlow in three aspects: framework generalization, correspondence generalization and cross-dataset generalization.

Time series forecasting is a relevant task that is performed in several real-world scenarios such as product sales analysis and prediction of energy demand. Given their accuracy performance, currently, Recurrent Neural Networks (RNNs) are the models of choice for this task. Despite their success in time series forecasting, less attention has been paid to make the RNNs trustworthy. For example, RNNs can not naturally provide an uncertainty measure to their predictions. This could be extremely useful in practice in several cases e.g. to detect when a prediction might be completely wrong due to an unusual pattern in the time series. Whittle Sum-Product Networks (WSPNs), prominent deep tractable probabilistic circuits (PCs) for time series, can assist an RNN with providing meaningful probabilities as uncertainty measure. With this aim, we propose RECOWN, a novel architecture that employs RNNs and a discriminant variant of WSPNs called Conditional WSPNs (CWSPNs). We also formulate a Log-Likelihood Ratio Score as better estimation of uncertainty that is tailored to time series and Whittle likelihoods. In our experiments, we show that RECOWNs are accurate and trustworthy time series predictors, able to "know when they do not know".

We propose a simple and effective strategy for data augmentation for low-resource machine reading comprehension (MRC). Our approach first pretrains the answer extraction components of a MRC system on the augmented data that contains approximate context of the correct answers, before training it on the exact answer spans. The approximate context helps the QA method components in narrowing the location of the answers. We demonstrate that our simple strategy substantially improves both document retrieval and answer extraction performance by providing larger context of the answers and additional training data. In particular, our method significantly improves the performance of BERT based retriever (15.12\%), and answer extractor (4.33\% F1) on TechQA, a complex, low-resource MRC task. Further, our data augmentation strategy yields significant improvements of up to 3.9\% exact match (EM) and 2.7\% F1 for answer extraction on PolicyQA, another practical but moderate sized QA dataset that also contains long answer spans.