Understanding the Power and Limitations of Teaching with Imperfect Knowledge

Devidze, Rati; Mansouri, Farnam; Haug, Luis; Chen, Yuxin; Singla, Adish

doi:10.24963/ijcai.2020/367

Cited by 7 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A machine teaching problem can be cast in a bi-level form where the upper-level problem defines the teacher's cost and the lower-level problem defines the learner's method. Variations of this bi-level form can be used to formulate teacher's optimization problem in a variety of learning settings, including supervised learning [46][47][48][49], imitation learning [50][51][52][53][54], and reinforcement learning [55][56][57][58][59]. In the proposed antidote data problem for clustering, the upper-level problem (teacher's cost) is the cost of adding antidote data, and the lower-level problem (learner) is the clustering algorithm.…”

Section: Results For Algorithmmentioning

confidence: 99%

Fair Clustering Using Antidote Data

Chhabra¹,

Singla²,

Mohapatra³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Clustering algorithms are widely utilized for many modern data science applications. This motivates the need to make outputs of clustering algorithms fair. Traditionally, new fair algorithmic variants to clustering algorithms are developed for specific notions of fairness. However, depending on the application context, different definitions of fairness might need to be employed. As a result, new algorithms and analysis need to be proposed for each combination of clustering algorithm and fairness definition. Additionally, each new algorithm would need to be reimplemented for deployment in a real-world system. Hence, we propose an alternate approach to fairness in clustering where we augment the original dataset with a small number of data points, called antidote data. When clustering is undertaken on this new dataset, the output is fair, for the chosen clustering algorithm and fairness definition. We formulate this as a general bi-level optimization problem which can accommodate any center-based clustering algorithms and fairness notions. We then categorize approaches for solving this bi-level optimization for different problem settings. Extensive experiments on different clustering algorithms and fairness notions show that our algorithms can achieve desired levels of fairness on many real-world datasets with a very small percentage of antidote data added. We also find that our algorithms achieve lower fairness costs and competitive clustering performance compared to other state-of-the-art fair clustering algorithms.Preprint. Under review.

show abstract

Section: Results For Algorithmmentioning

confidence: 99%

Fair Clustering Using Antidote Data

Chhabra¹,

Singla²,

Mohapatra³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, often excessive assumptions about the teacher's knowledge are made for deriving theoretical results. The teacher is usually assumed to have perfect knowledge about the computational model of students such as their learning rate and their background knowledge [31]. Thus, owing to the specificity of these theoretical models, they are difficult to apply in real-world teaching settings.…”

Section: Teaching Between the Machine And Humansmentioning

confidence: 99%

Explanatory machine learning for sequential human teaching

Ai¹,

Langer²,

Muggleton³

et al. 2022

Preprint

View full text Add to dashboard Cite

The topic of comprehensibility of machine-learned theories has recently drawn increasing attention. Inductive Logic Programming (ILP) uses logic programming to derive logic theories from small data based on abduction and induction techniques. Learned theories are represented in the form of rules as declarative descriptions of obtained knowledge. In earlier work, the authors provided the first evidence of a measurable increase in human comprehension based on machine-learned logic rules for simple classification tasks. In a later study, it was found that the presentation of machine-learned explanations to humans can produce both beneficial and harmful effects in the context of game learning. We continue our investigation of comprehensibility by examining the effects of the ordering of concept presentations on human comprehension. In this work, we examine the explanatory effects of curriculum order and the presence of machine-learned explanations for sequential problem-solving. We show that 1) there exist tasks A and B such that learning A before B has a better human comprehension with respect to learning B before A and 2) there exist tasks A and B such that the presence of explanations when learning A contributes to improved human comprehension when subsequently learning B. We propose a framework for the effects of sequential teaching on comprehension based on an existing definition of comprehensibility and provide evidence for support from data collected in human trials. Empirical results show that sequential teaching of concepts with increasing complexity a) has a beneficial effect on human comprehension and b) leads to human re-discovery of divide-and-conquer problemsolving strategies, and c) studying machine-learned explanations allows adaptations of human problem-solving strategy with better performance.

show abstract

“…Poisoning attacks is mathematically equivalent to the formulation of machine teaching with the teacher being the adversary (Goldman & Kearns, 1995;Zhu, 2015;Singla et al, 2014;Zhu et al, 2018;Chen et al, 2018;Mansouri et al, 2019;Peltola et al, 2019). A recent line of research has studied robust notions of teaching in settings where the teacher has limited information about the learner's dynamics (Dasgupta et al, 2019;Devidze et al, 2020;Cicalese et al, 2020), however, these works only consider supervised learning settings.…”

Section: Related Workmentioning

confidence: 99%

Reward Poisoning in Reinforcement Learning: Attacks Against Unknown Learners in Unknown Environments

Rakhsha¹,

Zhang²,

Zhu³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

We study black-box reward poisoning attacks against reinforcement learning (RL), in which an adversary aims to manipulate the rewards to mislead a sequence of RL agents with unknown algorithms to learn a nefarious policy in an environment unknown to the adversary a priori. That is, our attack makes minimum assumptions on the prior knowledge of the adversary: it has no initial knowledge of the environment or the learner, and neither does it observe the learner's internal mechanism except for its performed actions. We design a novel black-box attack, U2, that can provably achieve a near-matching performance to the state-of-the-art white-box attack, demonstrating the feasibility of reward poisoning even in the most challenging black-box setting.

show abstract

Understanding the Power and Limitations of Teaching with Imperfect Knowledge

Cited by 7 publications

References 0 publications

Fair Clustering Using Antidote Data

Fair Clustering Using Antidote Data

Explanatory machine learning for sequential human teaching

Reward Poisoning in Reinforcement Learning: Attacks Against Unknown Learners in Unknown Environments

Contact Info

Product

Resources

About