Users trust algorithms more when they can predict the algorithms’ behavior. Simple algorithms trivially yield predictively accurate mental models, but modern AI algorithms have often been assumed too complex for people to build predictive mental models, especially in the social media domain. In this paper, we describe conditions under which even complex algorithms can yield predictive mental models, opening up opportunities for a broader set of human-centered algorithms. We theorize that users will form an accurate predictive mental model of an algorithm’s behavior if and only if the algorithm simultaneously satisfies three criteria: (1) cognitive availability of the underlying concepts being modeled, (2) concept compactness (does it form a single cognitive construct?), and (3) high alignment between the person’s and algorithm’s execution of the concept. We evaluate this theory through a pre-registered experiment ( N = 1250) where users predict behavior of 25 social media feed ranking algorithms that vary on these criteria. We find that even complex (e.g., LLM-based) algorithms enjoy accurate prediction rates when they meet all criteria, and even simple (e.g., basic term count) algorithms fail to be predictable when a single criterion fails. We also find that these criteria determine outcomes beyond prediction accuracy, such as which mental models users deploy to make their predictions.

The widespread use of artificial intelligence requires finding energy-efficient paradigms for the field. We propose to reduce the energy consumption of Gaussian process regression using low-precision floating-point representations. We explore how low-precision representations impact the results of Gaussian process regression and how data set properties, implementation approach, model performance, and energy consumption interact. Our findings show that a well-conditioned kernel matrix allows reducing the energy consumption by up to 89.01% for 98.08% of arithmetic operations with little to no impact on model performance. Our findings are relevant whenever one needs to invert a symmetric full-rank matrix.

One of the biggest challenges to modern deep reinforcement learning (DRL) algorithms is sample efficiency. Many approaches learn a world model in order to train an agent entirely in imagination, eliminating the need for direct environment interaction during training. However, these methods often suffer from either a lack of imagination accuracy, exploration capabilities, or runtime efficiency. We propose HIEROS, a hierarchical policy that learns time abstracted world representations and imagines trajectories at multiple time scales in latent space. HIEROS uses an S5 layer-based world model, which predicts next world states in parallel during training and iteratively during environment interaction. Due to the special properties of S5 layers, our method can train in parallel and predict next world states iteratively during imagination. This allows for more efficient training than RNN-based world models and more efficient imagination than Transformer-based world models. We show that our approach outperforms the state of the art in terms of mean and median normalized human score on the Atari 100k benchmark, and that our proposed world model is able to predict complex dynamics very accurately. We also show that HIEROS displays superior exploration capabilities compared to existing approaches.

Variational Bayesian posterior inference often requires simplifying approximations such as mean-field parametrisation to ensure tractability. However, prior work has associated the variational mean-field approximation for Bayesian neural networks with underfitting in the case of small datasets or large model sizes. In this work, we show that invariances in the likelihood function of over-parametrised models contribute to this phenomenon because these invariances complicate the structure of the posterior by introducing discrete and/or continuous modes which cannot be well approximated by Gaussian mean-field distributions. In particular, we show that the mean-field approximation has an additional gap in the evidence lower bound compared to a purpose-built posterior that takes into account the known invariances. Importantly, this invariance gap is not constant; it vanishes as the approximation reverts to the prior. We proceed by first considering translation invariances in a linear model with a single data point in detail. We show that, while the true posterior can be constructed from a mean-field parametrisation, this is achieved only if the objective function takes into account the invariance gap. Then, we transfer our analysis of the linear model to neural networks. Our analysis provides a framework for future work to explore solutions to the invariance problem.

In this talk I will give an introduction into the field of machine learning and discuss why it is a crucial technology for Amazon.

Recently, learning deep models from dense data has received a lot of attention in tasks such as object recognition and signal processing. However, when dealing with non-sensory data about real-world entities, data is often sparse; for example people interaction with products in e-Commerce, people interacting with each other in social networks or word sequences in natural language. In this talk, I will share lessons learned over the past 10 years when learning predictive models based on sparse data: 1) how to scale the inference algorithms to distributed data setting, 2) how to automate the learning process by reducing the amount of hyper-parameters to zero, 3) how to deal with Zipf distributions when learning resource-constrained models, and 4) how to combine dense and sparse-learning algorithms. The talk will be drawing from many real-world experiences I gathered over the past decade in applications of the techniques in gaming, search, advertising and recommendations of systems developed at Microsoft, Facebook and Amazon.

Slow convergence and poor initial accuracy are two problems that plague efforts to use very large feature sets in online learning. This is especially true when only a few features are "active" in any training example, and the frequency of activations of different features is skewed. We show how these problems can be mitigated if a graph of relationships between features is known. We study this problem in a fully Bayesian setting, focusing on the problem of using Facebook user-IDs as features, with the social network giving the relationship structure. Our analysis uncovers significant problems with the obvious regularizations, and motivates a two-component mixture-model "social prior" that is provably better. Empirical results on large-scale click prediction problems show that our algorithm can learn as well as the baseline with 12M fewer training examples, and continuously outperforms it for over 60M examples. On a second problem using binned features, our model outperforms the baseline even after the latter sees 5x as much data.

Personalization is a ubiquitous phenomenon in our daily online experience. While such technology is critical for helping us combat the overload of information we face, in many cases, we may not even realize that our results are being tailored to our personal tastes and preferences. Worse yet, when such a system makes a mistake, we have little recourse to correct it.

The last ten years have seen a tremendous growth in Internet-based online services such as search, advertising, gaming and social networking. Today, it is important to analyze large collections of user interaction data as a first step in building predictive models for these services as well as learn these models in real-time. One of the biggest challenges in this setting is scale: not only does the sheer scale of data necessitate parallel processing but it also necessitates distributed models; with over 900 million active users at Facebook, any user-specific sets of features in a linear or non-linear model yields models of a size bigger than can be stored in a single system.

The prediction accuracy of any learning algorithm highly depends on the quality of the selected features; but often, the task of feature construction and selection is tedious and nonscalable. In recent years, however, there have been numerous projects with the goal of constructing general-purpose or domain-specific knowledge bases with entity-relationship-entity triples extracted from various Web sources or collected from user communities, e.g. YAGO, DBpedia, Freebase, UMLS, etc. This paper advocates the simple and yet far-reaching idea that the structured knowledge contained in such knowledge bases can be exploited to automatically extract features for general learning tasks. We introduce an expressive graph-based language for extracting features from such knowledge bases and a theoretical framework for constructing feature vectors from the extracted features. Our experimental evaluation on different learning scenarios provides evidence that the features derived through our framework can considerably improve the prediction accuracy, especially when the labeled data at hand is sparse.

Online video games can be seen as medium for the formation and maintenance of social relationships. In this paper, we explore what social relationships mean under the context of online First-Person Shooter (FPS) games, how these relationships influence game experience, and how players manage them. We combine qualitative interview and quantitative game log data, and find that despite the gap between the non-persistent game world and potentially persistent social relationships, a diversity of social relationships emerge and they play a central role in the enjoyment of online FPS games. We report the forms, development, and impact of such relationships, and discuss our findings in light of design implications and comparison with other game genres.

We consider the task of assigning experts from a portfolio of specialists in order to solve a set of tasks. We apply a Bayesian model which combines collaborative filtering with a feature-based description of tasks and experts to yield a general framework for managing a portfolio of experts. The model learns an embedding of tasks and problems into a latent space in which affinity is measured by the inner product. The model can be trained incrementally and can track non-stationary data, tracking potentially changing expert and task characteristics. The approach allows us to use a principled decision theoretic framework for expert selection, allowing the user to choose a utility function that best suits their objectives. The model component for taking into account the performance feedback data is pluggable, allowing flexibility. We apply the model to manage a portfolio of algorithms to solve hard combinatorial problems. This is a well studied area and we demonstrate a large improvement on the state of the art in one domain (constraint solving) and in a second domain (combinatorial auctions) created a portfolio that performed significantly better than any single algorithm.

We present a probabilistic model for generating personalised recommendations of items to users of a web service. The Matchbox system makes use of content information in the form of user and item meta data in combination with collaborative filtering information from previous user behavior in order to predict the value of an item for a user. Users and items are represented by feature vectors which are mapped into a low-dimensional `trait space' in which similarity is measured in terms of inner products. The model can be trained from different types of feedback in order to learn user-item preferences. Here we present three alternatives: direct observation of an absolute rating each user gives to some items, observation of a binary preference (like/ don't like) and observation of a set of ordinal ratings on a user-specific scale. Efficient inference is achieved by approximate message passing involving a combination of Expectation Propagation (EP) and Variational Message Passing. We also include a dynamics model which allows an item's popularity, a user's taste or a user's personal rating scale to drift over time. By using Assumed-Density Filtering (ADF) for training, the model requires only a single pass through the training data. This is an on-line learning algorithm capable of incrementally taking account of new data so the system can immediately reflect the latest user preferences. We evaluate the performance of the algorithm on the MovieLens and Netflix data sets consisting of approximately 1,000,000 and 100,000,000 ratings respectively. This demonstrates that training the model using the on-line ADF approach yields state-of-the-art performance with the option of improving performance further if computational resources are available by performing multiple EP passes over the training data.

No abstract available.

We apply probability theory to the task of proving whether a goal can be achieved by a player in an adversarial game. Such problems are solved by searching the game tree. We view this tree as a graphical model which yields a distribution over the (Boolean) outcome of the search before it terminates. Experiments show that a best-first search algorithm guided by this distribution explores a similar number of nodes as Proof-Number Search to solve Go problems. Knowledge is incorporated into search by using domain-specific models to provide prior distributions over the values of leaf nodes of the game tree. These are surrogate for the unexplored parts of the tree. The parameters of these models can be learned from previous search trees. Experiments on Go show that the speed of problem solving can be increased by orders of magnitude by this technique but care must be taken to avoid over-fitting.

We extend the Bayesian skill rating system TrueSkill to infer entire time series of skills of players by smoothing through time instead of (cid:12)ltering. The skill of each participating player, say, every year is represented by a latent skill variable which is a(cid:11)ected by the relevant game outcomes that year, and coupled with the skill variables of the previous and subsequent year. Inference in the resulting factor graph is carried out by approximate message passing (EP) along the time series of skills. As before the system tracks the uncertainty about player skills, explicitly models draws, can deal with any number of competing entities and can infer individual skills from team results. We extend the system to estimate player-speci(cid:12)c draw mar- gins. Based on these models we present an analysis of the skill curves of important players in the history of chess over the past 150 years. Results include plots of players’ lifetime skill development as well as the ability to compare the skills of di(cid:11)erent players across time. Our results indicate that a) the overall playing strength has increased over the past 150 years, and b) that modelling a player’s ability to force a draw provides signi(cid:12)cantly better predictive power.

We investigate the problem of learning to predict moves in the board game of Go from game records of expert players. In particular, we obtain a probability distribution over legal moves for professional play in a given position. This distribution has numerous applications in computer Go, including serving as an efficient stand-alone Go player. It would also be effective as a move selector and move sorter for game tree search and as a training tool for Go players. Our method has two major components: a) a pattern extraction scheme for efficiently harvesting patterns of given size and shape from expert game records and b) a Bayesian learning algorithm (in two variants) that learns a distribution over the values of a move given a board position based on the local pattern context. The system is trained on 181,000 expert games and shows excellent prediction performance as indicated by its ability to perfectly predict the moves made by professional Go players in 34% of test positions.

We study generalization properties of the area under the ROC curve (AUC), a quantity that has been advocated as an evaluation criterion for the bipartite ranking problem. The AUC is a different term than the error rate used for evaluation in classification problems; consequently, existing generalization bounds for the classification error rate cannot be used to draw conclusions about the AUC. In this paper, we define the expected accuracy of a ranking function (analogous to the expected error rate of a classification function), and derive distribution-free probabilistic bounds on the deviation of the empirical AUC of a ranking function (observed on a finite data sequence) from its expected accuracy. We derive both a large deviation bound, which serves to bound the expected accuracy of a ranking function in terms of its empirical AUC on a test sequence, and a uniform convergence bound, which serves to bound the expected accuracy of a learned ranking function in terms of its empirical AUC on a training sequence. Our uniform convergence bound is expressed in terms of a new set of combinatorial parameters that we term the bipartite rank-shatter coefficients; these play the same role in our result as do the standard VC-dimension related shatter coefficients (also known as the growth function) in uniform convergence results for the classification error rate. A comparison of our result with a recent uniform convergence result derived by Freund et al. (2003) for a quantity closely related to the AUC shows that the bound provided by our result can be considerably tighter.

We introduce two new functionals, the constrained covariance and the kernel mutual information, to measure the degree of independence of random variables. These quantities are both based on the covariance between functions of the random variables in reproducing kernel Hilbert spaces (RKHSs). We prove that when the RKHSs are universal, both functionals are zero if and only if the random variables are pairwise independent. We also show that the kernel mutual information is an upper bound near independence on the Parzen window estimate of the mutual information. Analogous results apply for two correlation-based dependence functionals introduced earlier: we show the kernel canonical correlation and the kernel generalised variance to be independence measures for universal kernels, and prove the latter to be an upper bound on the mutual information near independence. The performance of the kernel dependence functionals in measuring independence is verified in the context of independent component analysis.

The area under the ROC curve (AUC) has been advocated as an evalu- ation criterion for the bipartite ranking problem. We study large devi- ation properties of the AUC; in particular, we derive a distribution-free large deviation bound for the AUC which serves to bound the expected accuracy of a ranking function in terms of its empirical AUC on an inde- pendent test sequence. A comparison of our result with a corresponding large deviation result for the classification error rate suggests that the test sample size required to obtain an
-accurate estimate of the expected ac- curacy of a ranking function with δ-confidence is larger than that required to obtain an
-accurate estimate of the expected error rate of a classifi- cation function with the same confidence. A simple application of the union bound allows the large deviation bound to be extended to learned ranking functions chosen from finite function classes.

We introduce a new contrast function, the kernel mutual information (KMI), to measure the degree of independence of continuous random variables. This contrast function provides an approximate upper bound on the mutual information, as measured near independence, and is based on a kernel density estimate of the mutual information between a discretised approximation of the continuous random variables. We show that the kernel generalised variance (KGV) of F. Bach and M. Jordan (see JMLR, vol.3, p.1-48, 2002) is also an upper bound on the same kernel density estimate, but is looser. Finally, we suggest that the addition of a regularising term in the KGV causes it to approach the KMI, which motivates the introduction of this regularisation.

invariances with respect to given pattern Knowledge about local transformations can greatly improve the accuracy of classification. Previous approaches are either based on regularisation or on the gen- eration of virtual (transformed) examples. We develop a new frame- work for learning linear classifiers under known transformations based on semidefinite programming. We present a new learning algorithm— the Semidefinite Programming Machine (SDPM)—which is able to find a maximum margin hyperplane when the training examples are polynomial trajectories instead of single points. The solution is found to be sparse in dual variables and allows to identify those points on the trajectory with minimal real-valued output as virtual support vec- tors. Extensions to segments of trajectories, to more than one trans- formation parameter, and to learning with kernels are discussed. In experiments we use a Taylor expansion to locally approximate rota- tional invariance in pixel images from USPS and find improvements over known methods.

We present a modified version of the perceptron learning algorithm (PLA) which solves semidefinite programs (SDPs) in polynomial time. The algorithm is based on the following three observations: (i) Semidefinite programs are linear programs with infinitely many (linear) constraints; (ii) every linear program can be solved by a sequence of constraint satisfaction problems with linear constraints; (iii) in general, the perceptron learning algorithm solves a constraint satisfaction problem with linear constraints in finitely many updates. Combining the PLA with a probabilistic rescaling algorithm (which, on average, increases the size of the feasable region) results in a prob- abilistic algorithm for solving SDPs that runs in polynomial time. We present preliminary results which demonstrate that the algo- rithm works, but is not competitive with state-of-the-art interior point methods.

We present a framework for sparse Gaussian process (GP) methods which uses forward selection with criteria based on information- theoretic principles, previously suggested for active learning. Our goal is not only to learn d{sparse predictors (which can be evalu- ated in O(d) rather than O(n), d (cid:28) n, n the number of training points), but also to perform training under strong restrictions on time and memory requirements. The scaling of our method is at most O(n (cid:1) d2), and in large real-world classi(cid:12)cation experiments we show that it can match prediction performance of the popular support vector machine (SVM), yet can be signi(cid:12)cantly faster in training. In contrast to the SVM, our approximation produces esti- mates of predictive probabilities (‘error bars’), allows for Bayesian model selection and is less complex in implementation.

We present a bound on the generalization error of linear classifiers in terms of a refined margin quantity on the training sample. The result is obtained in a probably approximately correct (PAC)-Bayesian framework and is based on geometrical arguments in the space of linear classifiers. The new bound constitutes an exponential improvement of the so far tightest margin bound, which was developed in the luckiness framework, and scales logarithmically in the inverse margin. Even in the case of less training examples than input dimensions sufficiently large margins lead to nontrivial bound values and-for maximum margins-to a vanishing complexity term. In contrast to previous results, however, the new bound does depend on the dimensionality of feature space. The analysis shows that the classical margin is too coarse a measure for the essential quantity that controls the generalization error: the fraction of hypothesis space consistent with the training sample. The practical relevance of the result lies in the fact that the well-known support vector machine is optimal with respect to the new bound only if the feature vectors in the training sample are all of the same length. As a consequence, we recommend to use support vector machines (SVMs) on normalized feature vectors only. Numerical simulations support this recommendation and demonstrate that the new error bound can be used for the purpose of model selection.

In contrast to standard statistical learning theory which studies uniform bounds on the expected error we present a framework that exploits the specific learning algorithm used. Motivated by the luckiness framework [8] we are also able to exploit the serendipity of the training sample. The main difference to previous approaches lies in the complexity measure; rather than covering all hypothe(cid:173) ses in a given hypothesis space it is only necessary to cover the functions which could have been learned using the fixed learning algorithm. We show how the resulting framework relates to the VC, luckiness and compression frameworks. Finally, we present an application of this framework to the maximum margin algorithm for linear classifiers which results in a bound that exploits both the margin and the distribution of the data in feature space.

We present an algorithm that samples the hypothesis space of ker(cid:173) nel classifiers. Given a uniform prior over normalised weight vectors and a likelihood based on a model of label noise leads to a piece(cid:173) wise constant posterior that can be sampled by the kernel Gibbs sampler (KGS). The KGS is a Markov Chain Monte Carlo method that chooses a random direction in parameter space and samples from the resulting piecewise constant density along the line chosen. The KGS can be used as an analytical tool for the exploration of Bayesian transduction, Bayes point machines, active learning, and evidence-based model selection on small data sets that are contam(cid:173) inated with label noise. For a simple toy example we demonstrate experimentally how a Bayes point machine based on the KGS out(cid:173) performs an SVM that is incapable of taking into account label noise.

We present an improvement of Novikoff's perceptron convergence theorem. Reinterpreting this mistake bound as a margin dependent sparsity guarantee allows us to give a PAC-style generalisation er(cid:173) ror bound for the classifier learned by the perceptron learning algo(cid:173) rithm. The bound value crucially depends on the margin a support vector machine would achieve on the same data set using the same kernel. Ironically, the bound yields better guarantees than are cur(cid:173) rently available for the support vector solution itself.

We present a bound on the generalisation error of linear classifiers in terms of a refined margin quantity on the training set. The result is obtained in a PAC- Bayesian framework and is based on geometrical arguments in the space of linear classifiers. The new bound constitutes an exponential improvement of the so far tightest margin bound by Shawe-Taylor et al. [8] and scales logarithmically in the inverse margin. Even in the case of less training examples than input dimensions sufficiently large margins lead to non-trivial bound values and - plexity term. Furthermore, the classical margin is too coarse a measure for the essential quantity that controls the generalisation error: the volume ratio between the whole hypothesis space and the subset of consistent hypotheses. The practical relevance of the result lies in the fact that the well-known support vector machine is optimal w.r.t. the new bound only if the feature vectors are all of the same length. As a consequence we recommend to use SVMs on normalised feature vectors only - a recommendation that is well supported by our numerical experiments on two benchmark data sets. for maximum margins - to a vanishing com(cid:173)

also known as the Bayes point - The concept of averaging over classifiers is fundamental to the Bayesian analysis of learning. Based on this viewpoint, it has re(cid:173) cently been demonstrated for linear classifiers that the centre of mass of version space (the set of all classifiers consistent with the training set) - exhibits excel(cid:173) lent generalisation abilities. However, the billiard algorithm as pre(cid:173) sented in [4] is restricted to small sample size because it requires o (m 2 ) of memory and 0 (N . m2 ) computational steps where m is the number of training patterns and N is the number of random draws from the posterior distribution. In this paper we present a method based on the simple perceptron learning algorithm which allows to overcome this algorithmic drawback. The method is al(cid:173) gorithmically simple and is easily extended to the multi-class case. We present experimental results on the MNIST data set of hand(cid:173) written digits which show that Bayes point machines (BPMs) are competitive with the current world champion, the support vector machine. In addition, the computational complexity of BPMs can be tuned by varying the number of samples from the posterior. Finally, rejecting test points on the basis of their (approximative) posterior probability leads to a rapid decrease in generalisation er(cid:173) ror, e.g. 0.1% generalisation error for a given rejection rate of 10%.