Goto

Collaborating Authors

 behavior


ShoppingMMLU: AMassiveMulti-TaskOnline ShoppingBenchmarkforLargeLanguageModels

Neural Information Processing Systems

However,existingmodelsand benchmarks are commonly tailored to specific tasks, falling short of capturing the full complexity of online shopping. Large Language Models (LLMs), with their multi-task and few-shot learning abilities, have the potential to profoundly transform online shopping byalleviating task-specific engineering effortsandby providing users with interactiveconversations.


On the Periodic Behavior of Neural Network Training with Batch Normalization and Weight Decay

Neural Information Processing Systems

Training neural networks with batch normalization and weight decay has become a common practice in recent years. In this work, we show that their combined use may result in a surprising periodic behavior of optimization dynamics: the training process regularly exhibits destabilizations that, however, do not lead to complete divergence but cause a new period of training. We rigorously investigate the mechanism underlying the discovered periodic behavior from both empirical and theoretical points of view and analyze the conditions in which it occurs in practice. We also demonstrate that periodic behavior can be regarded as a generalization of two previously opposing perspectives on training with batch normalization and weight decay, namely the equilibrium presumption and the instability presumption.


Behavior From the Void: Unsupervised Active Pre-Training

Neural Information Processing Systems

We introduce a new unsupervised pre-training method for reinforcement learning called APT, which stands for Active Pre-Training. APT learns behaviors and representations by actively searching for novel states in reward-free environments. The key novel idea is to explore the environment by maximizing a non-parametric entropy computed in an abstract representation space, which avoids challenging density modeling and consequently allows our approach to scale much better in environments that have high-dimensional observations (e.g., image observations). We empirically evaluate APT by exposing task-specific reward after a long unsupervised pre-training phase. In Atari games, APT achieves human-level performance on 12 games and obtains highly competitive performance compared to canonical fully supervised RL algorithms. On DMControl suite, APT beats all baselines in terms of asymptotic performance and data efficiency and dramatically improves performance on tasks that are extremely difficult to train from scratch.


Reviews: Where Do You Think You're Going?: Inferring Beliefs about Dynamics from Behavior

Neural Information Processing Systems

The paper investigates the problem of inferring an agent's belief of the system dynamics of an MDP, given demonstrations of its behavior and the reward function it was optimizing. Knowledge of this internal belief can be used for Inverse Reinforcement Learning of an unknown task in the same environment. Furthermore, given the action provided by the agent, its intended action on the true dynamics can be inferred. This allows for assistive tele-operation, by applying the intended actions to the system instead of the provided ones. The proposed method models the agent using the model derived in maximum causal entropy inverse reinforcement learning.


Intellectual abilities of artificial intelligence (AI) - Semiwiki

#artificialintelligence

To understand AI’s capabilities and abilities we need to recognize the different components and subsets of AI. Terms like Neural Networks, Machine Learning (ML), and Deep Learning, need to be define and explained. In general, Artificial intelligence (AI) refers to the simulation of human intelligence in machines that are programmed to think like humans and…


A critique of pure learning and what artificial neural networks can learn from animal brains

#artificialintelligence

Not long after the invention of computers in the 1940s, expectations were high. Many believed that computers would soon achieve or surpass human-level intelligence. Herbert Simon, a pioneer of artificial intelligence (AI), famously predicted in 1965 that "machines will be capable, within twenty years, of doing any work a man can do"--to achieve general AI. Of course, these predictions turned out to be wildly off the mark. In the tech world today, optimism is high again.


Flexible Goal-Directed Agents' Behavior via DALI MASs and ASP Modules

AAAI Conferences

This paper describes the architecture that integrates DALI MASs (Multi-Agent Systems) and ASP (Answer Set Programming) modules for reaching goals in a flexible and timely way, where DALI is a computational-logic-based fully implemented agent-oriented logic programming language and ASP modules includes solvers that allow affordable and flexible planning capabilities. The proposed DALI MAS architecture exploits such modules for flexible goal decomposition and planning, with the possibility to select plans according to a suite of possible preferences and to re-plan upon need. We present an abstract case-study concerning DALI agents which cooperate for exploring an unknown territory under changing circumstances in an optimal or at least sub-optimal fashion. The architecture can be exploited not only by DALI agents, but rather by any kind of logical agent.


On the Behavior of Convolutional Nets for Feature Extraction

Journal of Artificial Intelligence Research

Deep neural networks are representation learning techniques. During training, a deep net is capable of generating a descriptive language of unprecedented size and detail in machine learning. Extracting the descriptive language coded within a trained CNN model (in the case of image data), and reusing it for other purposes is a field of interest, as it provides access to the visual descriptors previously learnt by the CNN after processing millions of images, without requiring an expensive training phase. Contributions to this field (commonly known as feature representation transfer or transfer learning) have been purely empirical so far, extracting all CNN features from a single layer close to the output and testing their performance by feeding them to a classifier. This approach has provided consistent results, although its relevance is limited to classification tasks. In a completely different approach, in this paper we statistically measure the discriminative power of every single feature found within a deep CNN, when used for characterizing every class of 11 datasets. We seek to provide new insights into the behavior of CNN features, particularly the ones from convolutional layers, as this can be relevant for their application to knowledge representation and reasoning. Our results confirm that low and middle level features may behave differently to high level features, but only under certain conditions. We find that all CNN features can be used for knowledge representation purposes both by their presence or by their absence, doubling the information a single CNN feature may provide. We also study how much noise these features may include, and propose a thresholding approach to discard most of it. All these insights have a direct application to the generation of CNN embedding spaces.


Relating Children’s Automatically Detected Facial Expressions to Their Behavior in RoboTutor

AAAI Conferences

Can student behavior be anticipated in real-time so that an intelligent tutor system can adapt its content to keep the student engaged? Current methods detect affective states of students during learning session to determine their engagement levels but apply the learning in next session in the form of intervention policies and tutor responses. However, if students' imminent behavioral action could be anticipated from their affective states in real-time, this could lead to much more responsive intervention policies by the tutor and assist in keeping the student engaged in an activity, thereby increasing tutor efficacy as well as student engagement levels. In this paper we explore if there exist any links between a student's affective states and his/her imminent behavior action in RoboTutor, an intelligent tutor system for children to learn math, reading and writing. We then exploit our findings to develop a real-time student behavior prediction module.


How AI can help you stay ahead of cybersecurity threats

#artificialintelligence

Since the 2013 Target breach, it's been clear that companies need to respond better to security alerts even as volumes have gone up. With this year's fast-spreading ransomware attacks and ever-tightening compliance requirements, response must be much faster. Adding staff is tough with the cybersecurity hiring crunch, so companies are turning to machine learning and artificial intelligence (AI) to automate tasks and better detect bad behavior. In a cybersecurity context, AI is software that perceives its environment well enough to identify events and take action against a predefined purpose. AI is particularly good at recognizing patterns and anomalies within them, which makes it an excellent tool to detect threats.