Helsinki Institute for Information Technology

About: Helsinki Institute for Information Technology is a facility organization based out in Espoo, Finland. It is known for research contribution in the topics: Population & Bayesian network. The organization has 630 authors who have published 1962 publications receiving 63426 citations.

Synchronous Counting and Computational Algorithm Design

Abstract: Consider a complete communication network on n nodes, each of which is a state machine with s states. In synchronous 2-counting, the nodes receive a common clock pulse and they have to agree on which pulses are "odd" and which are "even". We require that the solution is self-stabilising (reaching the correct operation from any initial state) and it tolerates f Byzantine failures (nodes that send arbitrary misinformation). Prior algorithms are expensive to implement in hardware: they require a source of random bits or a large number of statesas. We use computational techniques to construct very compact deterministic algorithms for the first non-trivial case of f = 1. While no algorithm exists for n < 4, we show that as few as 3 states are sufficient for all values n ≤ 4. We prove that the problem cannot be solved with only 2 states for n = 4, but there is a 2-state solution for all values n ≤ 6.

Controlled permutations for testing adaptive learning models

Abstract: We study evaluation of supervised learning models that adapt to changing data distribution over time (concept drift). The standard testing procedure that simulates online arrival of data (test-then-train) may not be sufficient to generalize about the performance, since that single test concludes how well a model adapts to this fixed configuration of changes, while the ultimate goal is to assess the adaptation to changes that happen unexpectedly. We propose a methodology for obtaining datasets for multiple tests by permuting the order of the original data. A random permutation is not suitable, as it makes the data distribution uniform over time and destroys the adaptive learning task. Therefore, we propose three controlled permutation techniques that make it possible to acquire new datasets by introducing restricted variations in the order of examples. The control mechanisms with theoretical guarantees of preserving distributions ensure that the new sets represent close variations of the original learning task. Complementary tests on such sets allow to analyze sensitivity of the performance to variations in how changes happen and this way enrich the assessment of adaptive supervised learning models.

Data transmission and base-station placement for optimizing the lifetime of wireless sensor networks

Abstract: In this paper, we study the fundamental optimization problem in wireless sensor networks of base-station positioning such that data from the sensors may be transmitted to it in an energy-efficient manner. We primarily consider the setting where a sensor transmits all of its data directly to the base-station or relays it via one other node. This setting provides two benefits: low duty-cycling due to limited synchronization requirements between nodes and low end-to-end delay due to the limited number of hops in the routes. Given the battery limitations of the sensor nodes, our objective is to maximize the network lifetime. First, we present efficient algorithms for computing a transmission scheme for the sensors given a fixed base-station and show how to implement these in a distributed fashion with only a constant number of messages per sensor. Next, we show that the optimization problem for the setting where sensors may transmit data through more than 2 hops is NP-Hard. Finally, we present efficient algorithms for the problem of locating the base-station and simultaneously finding a transmission scheme. We compare our algorithms with linear-programming based algorithms for more general settings through extensive simulations and outline the benefits of the different approaches.

Active Learning for Decision-Making from Imbalanced Observational Data

Abstract: Machine learning can help personalized decision support by learning models to predict individual treatment effects (ITE). This work studies the reliability of prediction-based decision-making in a task of deciding which action $a$ to take for a target unit after observing its covariates $\tilde{x}$ and predicted outcomes $\hat{p}(\tilde{y} \mid \tilde{x}, a)$. An example case is personalized medicine and the decision of which treatment to give to a patient. A common problem when learning these models from observational data is imbalance, that is, difference in treated/control covariate distributions, which is known to increase the upper bound of the expected ITE estimation error. We propose to assess the decision-making reliability by estimating the ITE model's Type S error rate, which is the probability of the model inferring the sign of the treatment effect wrong. Furthermore, we use the estimated reliability as a criterion for active learning, in order to collect new (possibly expensive) observations, instead of making a forced choice based on unreliable predictions. We demonstrate the effectiveness of this decision-making aware active learning in two decision-making tasks: in simulated data with binary outcomes and in a medical dataset with synthetic and continuous treatment outcomes.

The semiotics of the message and the messenger: How nonverbal communication affects fairness perception.

Abstract: Nonverbal communication determines much of how we perceive explicit, verbal messages. Facial expressions and social touch, for example, influence affinity and conformity. To understand the interaction between nonverbal and verbal information, we studied how the psychophysiological time-course of semiotics-the decoding of the meaning of a message-is altered by interpersonal touch and facial expressions. A virtual-reality-based economic decision-making game, ultimatum, was used to investigate how participants perceived, and responded to, financial offers of variable levels of fairness. In line with previous studies, unfair offers evoked medial frontal negativity (MFN) within the N2 time window, which has been interpreted as reflecting an emotional reaction to violated social norms. Contrary to this emotional interpretation of the MFN, however, nonverbal signals did not modulate the MFN component, only affecting fairness perception during the P3 component. This suggests that the nonverbal context affects the late, but not the early, stage of fairness perception. We discuss the implications of the semiotics of the message and the messenger as a process by which parallel information sources of "who says what" are integrated in reverse order: of the message, then the messenger.