Deokgun Park

Deokgun Park, Md Ashaduzzaman Rubel Mondol, SM Mazharul Islam et al.

The neural implementation of operant conditioning with few trials is unclear. We propose a Hippocampus-Inspired Cognitive Architecture (HICA) as a neural mechanism for operant conditioning. HICA explains a learning mechanism in which agents can learn a new behavior policy in a few trials, as mammals do in operant conditioning experiments. HICA is composed of two different types of modules. One is a universal learning module type that represents a cortical column in the neocortex gray matter. The working principle is modeled as Modulated Heterarchical Prediction Memory (mHPM). In mHPM, each module learns to predict a succeeding input vector given the sequence of the input vectors from lower layers and the context vectors from higher layers. The prediction is fed into the lower layers as a context signal (top-down feedback signaling), and into the higher layers as an input signal (bottom-up feedforward signaling). Rewards modulate the learning rate in those modules to memorize meaningful sequences effectively. In mHPM, each module updates in a local and distributed way compared to conventional end-to-end learning with backpropagation of the single objective loss. This local structure enables the heterarchical network of modules. The second type is an innate, special-purpose module representing various organs of the brain's subcortical system. Modules modeling organs such as the amygdala, hippocampus, and reward center are pre-programmed to enable instinctive behaviors. The hippocampus plays the role of the simulator. It is an autoregressive prediction model of the top-most level signal with a loop structure of memory, while cortical columns are lower layers that provide detailed information to the simulation. The simulation becomes the basis for learning with few trials and the deliberate planning required for operant conditioning.

Deokgun Park

In this paper, we present our research on programming human-level artificial intelligence (HLAI), including 1) a definition of HLAI, 2) an environment to develop and test HLAI, and 3) a cognitive architecture for HLAI. The term AI is used in a broad meaning, and HLAI is not clearly defined. I claim that the essence of Human-Level Intelligence to be the capability to learn from others' experiences via language. The key is that the event described by language has the same effect as if the agent experiences it firsthand for the update of the behavior policy. To develop and test models with such a capability, we are developing a simulated environment called SEDRo. There is a 3D Home, and a mother character takes care of the baby (the learning agent) and teaches languages. The environment provides comparable experiences to that of a human baby from birth to one year. Finally, I propose a cognitive architecture of HLAI called Modulated Heterarchical Prediction Memory (mHPM). In mHPM, there are three components: a universal module that learns to predict the next vector given the sequence of vector signals, a heterarchical network of those modules, and a reward-based modulation of learning. mHPM models the workings of the neocortex but the innate auxiliary units such hippocampus, reward system, instincts, and amygdala play critical roles, too.

Md Ashaduzzaman Rubel Mondol, Aishwarya Pothula, Deokgun Park

Task-specific AI agents are showing remarkable performance across different domains. But modeling generalized AI agents like human intelligence will require more than current datasets or only reward-based environments that don't include experiences that an infant gathers throughout its initial stages. In this paper, we present Simulated Environment for Developmental Robotics (SEDRo). It simulates the environments for a baby agent that a human baby experiences throughout the pre-born fetus stage to post-birth 12 months. SEDRo also includes a mother character to provide social interaction with the agent. To evaluate different developmental milestones of the agent, SEDRo incorporates some experiments from developmental psychology.

Deokgun Park, Md Ashaduzzaman Rubel Mondol, Aishwarya Pothula et al.

Despite recent advances of AI research in many application-specific domains, we do not know how to build a human-level artificial intelligence (HLAI). We conjecture that learning from others' experience with the language is the essential characteristic that distinguishes human intelligence from the rest. Humans can update the action-value function with the verbal description as if they experience states, actions, and corresponding rewards sequences firsthand. In this paper, we present a classification of intelligence according to how individual agents learn and propose a definition and a test for HLAI. The main idea is that language acquisition without explicit rewards can be a sufficient test for HLAI.

Aishwarya Pothula, Md Ashaduzzaman Rubel Mondol, Sanath Narasimhan et al.

Even with impressive advances in application-specific models, we still lack knowledge about how to build a model that can learn in a human-like way and do multiple tasks. To learn in a human-like way, we need to provide a diverse experience that is comparable to humans. In this paper, we introduce our ongoing effort to build a simulated environment for developmental robotics (SEDRo). SEDRo provides diverse human experiences ranging from those of a fetus to a 12th-month-old. A series of simulated tests based on developmental psychology will be used to evaluate the progress of a learning model. We anticipate SEDRo to lower the cost of entry and facilitate research in the developmental robotics community.

SM Mazharul Islam, Md Ashaduzzaman Rubel Mondol, Aishwarya Pothula et al.

As the current trend of artificial intelligence is shifting towards self-supervised learning, conventional norms such as highly curated domain-specific data, application-specific learning models, extrinsic reward based learning policies etc. might not provide with the suitable ground for such developments. In this paper, we introduce SEDRo, a Simulated Environment for Developmental Robotics which allows a learning agent to have similar experiences that a human infant goes through from the fetus stage up to 12 months. A series of simulated tests based on developmental psychology will be used to evaluate the progress of a learning model.

Amira Chalbi, Jacob Ritchie, Deokgun Park et al.

The Law of Common Fate from Gestalt psychology states that visual objects moving with the same velocity along parallel trajectories will be perceived by a human observer as grouped. However, the concept of common fate is much broader than mere velocity; in this paper we explore how common fate results from coordinated changes in luminance and size. We present results from a crowdsourced graphical perception study where we asked workers to make perceptual judgments on a series of trials involving four graphical objects under the influence of conflicting static and dynamic visual factors (position, size and luminance) used in conjunction. Our results yield the following rankings for visual grouping: motion > (dynamic luminance, size, luminance); dynamic size > (dynamic luminance, position); and dynamic luminance > size. We also conducted a follow-up experiment to evaluate the three dynamic visual factors in a more ecologically valid setting, using both a Gapminder-like animated scatterplot and a thematic map of election data. The results indicate that in practice the relative grouping strengths of these factors may depend on various parameters including the visualization characteristics and the underlying data. We discuss design implications for animated transitions in data visualization.

Deokgun Park, Sung-Hee Kim, Niklas Elmqvist

Overplotting of data points is a common problem when visualizing large datasets in a scatterplot, particularly when mapping nominal dimensions to one of the scatterplot axes. Transparency, aggregation, and jittering have previously been suggested to address this issue, but these solutions all have drawbacks for assessing the data distribution in the plot. We propose gatherplots, an extension of scatterplots that eliminates overplotting, particularly for nominal variables. In gatherplots, every data point that maps to the same position coalesces to form a stacked entity, thereby making it easier to compare the absolute and relative sizes of data groupings. The size and aspect ratio of data points can also be changed dynamically to make it easier to compare the composition of different groups. Furthermore, several embedded interaction techniques support slicing and dicing the gatherplot by pivoting on particular dimensions, ranges, and values in the dataset. Our evaluation shows that gatherplots enable users from the general public to judge the relative portion of subgroups more quickly and more correctly than when using conventional scatterplots with jittering. Furthermore, a review conducted by a group of visualization experts evaluated and commented on the gatherplot design.