Lottery Tickets in Linear Models: An Analysis of Iterative Magnitude Pruning

We analyse the pruning procedure behind the lottery ticket hypothesis arXiv:1803.03635v5 [cs.LG], iterative magnitude pruning (IMP), when applied to linear models trained by gradient flow. We begin by presenting sufficient conditions on the statistical structure of the features, under which IMP prunes those features that have smallest projection onto the data. Following this, we explore IMP as a method for sparse estimation and sparse prediction in noisy settings, with minimal assumptions on the design matrix. The same techniques are then applied to derive corresponding results for threshold pruning. Finally, we present experimental evidence of the regularising effect of IMP. We hope that our work will contribute to a theoretically grounded understanding of lottery tickets and how they emerge from IMP.

How benign is benign overfitting?

We investigate two causes for adversarial vulnerability in deep neural networks: bad data and (poorly) trained models. When trained with SGD, deep neural networks essentially achieve zero training error, even in the presence of label noise, while also exhibiting good generalization on natural test data, something referred to as benign overfitting [2, 10]. However, these models are vulnerable to adversarial attacks. We identify label noise as one of the causes for adversarial vulnerability, and provide theoretical and empirical evidence in support of this. Surprisingly, we find several instances of label noise in datasets such as MNIST and CIFAR, and that robustly trained models incur training error on some of these, i.e. they don't fit the noise. However, removing noisy labels alone does not suffice to achieve adversarial robustness. Standard training procedures bias neural networks towards learning "simple" classification boundaries, which may be less robust than more complex ones. We observe that adversarial training does produce more complex decision boundaries. We conjecture that in part the need for complex decision boundaries arises from sub-optimal representation learning. By means of simple toy examples, we show theoretically how the choice of representation can drastically affect adversarial robustness.

Differentiable Causal Backdoor Discovery

Discovering the causal effect of a decision is critical to nearly all forms of decision-making. In particular, it is a key quantity in drug development, in crafting government policy, and when implementing a real-world machine learning system. Given only observational data, confounders often obscure the true causal effect. Luckily, in some cases, it is possible to recover the causal effect by using certain observed variables to adjust for the effects of confounders. However, without access to the true causal model, finding this adjustment requires brute-force search. In this work, we present an algorithm that exploits auxiliary variables, similar to instruments, in order to find an appropriate adjustment by a gradient-based optimization method. We demonstrate that it outperforms practical alternatives in estimating the true causal effect, without knowledge of the full causal graph.

The Statistical Complexity of Early Stopped Mirror Descent

Recently there has been a surge of interest in understanding implicit regularization properties of iterative gradient-based optimization algorithms. In this paper, we study the statistical guarantees on the excess risk achieved by early stopped unconstrained mirror descent algorithms applied to the unregularized empirical risk with squared loss for linear models and kernel methods. We identify a link between offset Rademacher complexities and potential-based analysis of mirror descent that allows disentangling statistics from optimization in the analysis of such algorithms. Our main result characterizes the statistical performance of the path traced by the iterates of mirror descent in terms of offset complexities of certain function classes depending only on the choice of the mirror map, initialization point, step-size, and number of iterations. We apply our theory to recover, in a rather clean and elegant manner, some of the recent results in the implicit regularization literature, while also showing how to improve upon them in some settings.

Online k-means Clustering

We study the problem of online clustering where a clustering algorithm has to assign a new point that arrives to one of $k$ clusters. The specific formulation we use is the $k$-means objective: At each time step the algorithm has to maintain a set of k candidate centers and the loss incurred is the squared distance between the new point and the closest center. The goal is to minimize regret with respect to the best solution to the $k$-means objective ($\mathcal{C}$) in hindsight. We show that provided the data lies in a bounded region, an implementation of the Multiplicative Weights Update Algorithm (MWUA) using a discretized grid achieves a regret bound of $\tilde{O}(\sqrt{T})$ in expectation. We also present an online-to-offline reduction that shows that an efficient no-regret online algorithm (despite being allowed to choose a different set of candidate centres at each round) implies an offline efficient algorithm for the $k$-means problem. In light of this hardness, we consider the slightly weaker requirement of comparing regret with respect to $(1 + \epsilon) \mathcal{C}$ and present a no-regret algorithm with runtime $O\left(T(\mathrm{poly}(log(T),k,d,1/\epsilon)^{k(d+O(1))}\right)$. Our algorithm is based on maintaining an incremental coreset and an adaptive variant of the MWUA. We show that na\"{i}ve online algorithms, such as \emph{Follow The Leader}, fail to produce sublinear regret in the worst case. We also report preliminary experiments with synthetic and real-world data.

On the Hardness of Robust Classification

It is becoming increasingly important to understand the vulnerability of machine learning models to adversarial attacks. In this paper we study the feasibility of robust learning from the perspective of computational learning theory, considering both sample and computational complexity. In particular, our definition of robust learnability requires polynomial sample complexity. We start with two negative results. We show that no non-trivial concept class can be robustly learned in the distribution-free setting against an adversary who can perturb just a single input bit. We show moreover that the class of monotone conjunctions cannot be robustly learned under the uniform distribution against an adversary who can perturb $\omega(\log n)$ input bits. However if the adversary is restricted to perturbing $O(\log n)$ bits, then the class of monotone conjunctions can be robustly learned with respect to a general class of distributions (that includes the uniform distribution). Finally, we provide a simple proof of the computational hardness of robust learning on the boolean hypercube. Unlike previous results of this nature, our result does not rely on another computational model (e.g. the statistical query model) nor on any hardness assumption other than the existence of a hard learning problem in the PAC framework.

Implicit Regularization for Optimal Sparse Recovery

We investigate implicit regularization schemes for gradient descent methods applied to unpenalized least squares regression to solve the problem of reconstructing a sparse signal from an underdetermined system of linear measurements under the restricted isometry assumption. For a given parametrization yielding a non-convex optimization problem, we show that prescribed choices of initialization, step size and stopping time yield a statistically and computationally optimal algorithm that achieves the minimax rate with the same cost required to read the data up to poly-logarithmic factors. Beyond minimax optimality, we show that our algorithm adapts to instance difficulty and yields a dimension-independent rate when the signal-to-noise ratio is high enough. Key to the computational efficiency of our method is an increasing step size scheme that adapts to refined estimates of the true solution. We validate our findings with numerical experiments and compare our algorithm against explicit $\ell_{1}$ penalization. Going from hard instances to easy ones, our algorithm is seen to undergo a phase transition, eventually matching least squares with an oracle knowledge of the true support.

Adaptive Reduced Rank Regression

Low rank regression has proven to be useful in a wide range of forecasting problems. However, in settings with a low signal-to-noise ratio, it is known to suffer from severe overfitting. This paper studies the reduced rank regression problem and presents algorithms with provable generalization guarantees. We use adaptive hard rank-thresholding in two different parts of the data analysis pipeline. First, we consider a low rank projection of the data to eliminate the components that are most likely to be noisy. Second, we perform a standard multivariate linear regression estimator on the data obtained in the first step, and subsequently consider a low-rank projection of the obtained regression matrix. Both thresholding is performed in a data-driven manner and is required to prevent severe overfitting as our lower bounds show. Experimental results show that our approach either outperforms or is competitive with existing baselines.

Decentralized Cooperative Stochastic Multi-armed Bandits

We study a decentralized cooperative stochastic multi-armed bandit problem with $K$ arms on a network of $N$ agents. In our model, the reward distribution of each arm is agent-independent. Each agent chooses iteratively one arm to play and then communicates to her neighbors. The aim is to minimize the total network regret. We design a fully decentralized algorithm that uses a running consensus procedure to compute, with some delay, accurate estimations of the average of rewards obtained by all the agents for each arm, and then uses an upper confidence bound algorithm that accounts for the delay and error of the estimations. We analyze the algorithm and up to a constant our regret bounds are better for all networks than other algorithms designed to solve the same problem. For some graphs, our regret bounds are significantly better.

Statistical Windows in Testing for the Initial Distribution of a Reversible Markov Chain

We study the problem of hypothesis testing between two discrete distributions, where we only have access to samples after the action of a known reversible Markov chain, playing the role of noise. We derive instance-dependent minimax rates for the sample complexity of this problem, and show how its dependence in time is related to the spectral properties of the Markov chain. We show that there exists a wide statistical window, in terms of sample complexity for hypothesis testing between different pairs of initial distributions. We illustrate these results in several concrete examples.